research - fzab.si › uploads › file › kumar, r_, 2011... · step viii: writing a research...
TRANSCRIPT
RESEARCHMETHODOLOGYastep-by-stepguideforbeginners
SAGEhasbeenpartof theglobalacademiccommunitysince1965,supportinghighquality researchandlearningthattransformssocietyandourunderstandingofindividuals,groups,andcultures.SAGEis the independent, innovative, natural home for authors, editors and societies who share ourcommitmentandpassionforthesocialsciences.
Findoutmoreat:www.sagepublications.com
3rdedition
RESEARCHMETHODOLOGYastep-by-stepguideforbeginners
RanjitKumar
©RanjitKumar1999,2005,2011Firsteditionpublished1999Secondeditionpublished2005.Reprinted2007,2008(twice),2009(twice)Thisthirdeditionpublished2011
Apartfromanyfairdealingforthepurposesofresearchorprivatestudy,orcriticismorreview,aspermittedundertheCopyright,DesignsandPatentsAct,1988,thispublicationmaybereproduced,storedortransmittedinanyform,orbyanymeans,onlywiththepriorpermissioninwritingofthepublishers,orinthecaseofreprographicreproduction,inaccordancewiththetermsoflicencesissuedbytheCopyrightLicensingAgency.Enquiriesconcerningreproductionoutsidethosetermsshouldbesenttothepublishers.
SAGEPublicationsLtd1Oliver’sYard55CityRoadLondonEC1Y1SP
SAGEPublicationsInc.2455TellerRoadThousandOaks,California91320
SAGEPublicationsIndiaPvtLtdB1/I1MohanCooperativeIndustrialAreaMathuraRoadNewDelhi110044
SAGEPublicationsAsia-PacificPteLtd33PekinStreet#02-01FarEastSquareSingapore048763
LibraryofCongressControlNumberavailable
BritishLibraryCataloguinginPublicationdata
AcataloguerecordforthisbookisavailablefromtheBritishLibrary
ISBN978-1-84920-300-5ISBN978-1-84920-301-2(pbk)
TypesetbyC&MDigitals(P)Ltd,Chennai,IndiaPrintedandboundinGreatBritainbyTJInternationalLtd,Padstow,CornwallPrintedonpaperfromsustainableresources
Tomydaughter,Parul
Contents
ListoffiguresListoftablesPreface
1Research:awayofthinkingResearch:anintegralpartofyourpracticeResearch:awaytogatherevidenceforyourpracticeApplicationsofresearchResearch:whatdoesitmean?Theresearchprocess:characteristicsandrequirementsTypesofresearchTypesofresearch:applicationperspectiveTypesofresearch:objectivesperspectiveTypesofresearch:modeofenquiryperspectiveParadigmsofresearchSummary
2Theresearchprocess:aquickglanceTheresearchprocess:aneight-stepmodelPhaseI:decidingwhattoresearchStepI:formulatingaresearchproblemPhaseII:planningaresearchstudyStepII:conceptualisingaresearchdesignStepIII:constructinganinstrumentfordatacollectionStepIV:selectingasampleStepV:writingaresearchproposalPhaseIII:conductingaresearchstudyStepVI:collectingdataStepVII:processinganddisplayingdataStepVIII:writingaresearchreportSummary
STEPIFORMULATINGARESEARCHPROBLEM
3ReviewingtheliteratureTheplaceoftheliteraturereviewinresearchBringingclarityandfocustoyourresearchproblemImprovingyourresearchmethodologyBroadeningyourknowledgebaseinyourresearchareaEnablingyoutocontextualiseyourfindings
HowtoreviewtheliteratureSearchingfortheexistingliteratureReviewingtheselectedliteratureDevelopingatheoreticalframeworkDevelopingaconceptualframeworkWritingabouttheliteraturereviewedSummary
4FormulatingaresearchproblemTheresearchproblemTheimportanceofformulatingaresearchproblemSourcesofresearchproblemsConsiderationsinselectingaresearchproblemStepsinformulatingaresearchproblemTheformulationofresearchobjectivesThestudypopulationEstablishingoperationaldefinitionsFormulatingaresearchprobleminqualitativeresearchSummary
5IdentifyingvariablesWhatisavariable?ThedifferencebetweenaconceptandavariableConvertingconceptsintovariablesTypesofvariableFromtheviewpointofcausalrelationshipFromtheviewpointofthestudydesignFromtheviewpointoftheunitofmeasurementTypesofmeasurementscaleThenominalorclassificatoryscaleTheordinalorrankingscaleTheintervalscaleTheratioscaleSummary
6ConstructinghypothesesThedefinitionofahypothesisThefunctionsofahypothesisThetestingofahypothesisThecharacteristicsofahypothesisTypesofhypothesisErrorsintestingahypothesisHypothesesinqualitativeresearchSummary
STEPIICONCEPTUALISINGARESEARCHDESIGN
7TheresearchdesignWhatisaresearchdesign?ThefunctionsofaresearchdesignThetheoryofcausalityandtheresearchdesignSummary
8SelectingastudydesignDifferencesbetweenquantitativeandqualitativestudydesignsStudydesignsinquantitativeresearchStudydesignsbasedonthenumberofcontactsStudydesignsbasedonthereferenceperiodStudydesignsbasedonthenatureoftheinvestigationOtherdesignscommonlyusedinquantitativeresearchThecross-overcomparativeexperimentaldesignThereplicatedcross-sectionaldesignTrendstudiesCohortstudiesPanelstudiesBlindstudiesDouble-blindstudiesStudydesignsinqualitativeresearchCasestudyOralhistoryFocusgroups/groupinterviewsParticipantobservationHolisticresearchCommunitydiscussionforumsReflectivejournallogOthercommonlyusedphilosophy-guideddesignsActionresearchFeministresearchParticipatoryandcollaborativeresearchenquirySummary
STEPIIICONSTRUCTINGANINSTRUMENTFORDATACOLLECTION
9SelectingamethodofdatacollectionDifferencesinthemethodsofdatacollectioninquantitativeandqualitativeresearchMajorapproachestoinformationgatheringCollectingdatausingprimarysourcesObservationTheinterviewThequestionnaireConstructingaresearchinstrumentinquantitativeresearchAskingpersonalandsensitivequestionsTheorderofquestionsPre-testingaresearchinstrument
PrerequisitesfordatacollectionMethodsofdatacollectioninqualitativeresearchConstructingaresearchinstrumentinqualitativeresearchCollectingdatausingsecondarysourcesProblemswithusingdatafromsecondarysourcesSummary
10CollectingdatausingattitudinalscalesMeasurementofattitudesinquantitativeandqualitativeresearchAttitudinalscalesinquantitativeresearchFunctionsofattitudinalscalesDifficultiesindevelopinganattitudinalscaleTypesofattitudinalscaleThesummatedratingorLikertscaleTheequal-appearingintervalorThurstonescaleThecumulativeorGuttmanscaleAttitudinalscalesandmeasurementscalesAttitudesandqualitativeresearchSummary
11EstablishingthevalidityandreliabilityofaresearchinstrumentTheconceptofvalidityTypesofvalidityinquantitativeresearchFaceandcontentvalidityConcurrentandpredictivevalidityConstructvalidityTheconceptofreliabilityFactorsaffectingthereliabilityofaresearchinstrumentMethodsofdeterminingthereliabilityofaninstrumentinquantitativeresearchExternalconsistencyproceduresInternalconsistencyproceduresValidityandreliabilityinqualitativeresearchSummarySTEPIVSELECTINGASAMPLE
12SelectingasampleThedifferencesbetweensamplinginquantitativeandqualitativeresearchSamplinginquantitativeresearchTheconceptofsamplingSamplingterminologyPrinciplesofsamplingFactorsaffectingtheinferencesdrawnfromasampleAimsinselectingasampleTypesofsamplingNon-random/non-probabilitysamplingdesignsinquantitativeresearchSystematicsamplingdesign:a‘mixed’design
ThecalculationofsamplesizeSamplinginqualitativeresearchTheconceptofsaturationpointinqualitativeresearchSummarySTEPVWRITINGARESEARCHPROPOSAL
13HowtowritearesearchproposalTheresearchproposalinquantitativeandqualitativeresearchContentsofaresearchproposalPreamble/introductionTheproblemObjectivesofthestudyHypothesestobetestedStudydesignThesettingMeasurementproceduresEthicalissuesSamplingAnalysisofdataStructureofthereportProblemsandlimitationsAppendixWorkscheduleSummarySTEPVICOLLECTINGDATA
14ConsideringethicalissuesindatacollectionEthics:theconceptStakeholdersinresearchEthicalissuestoconsiderconcerningresearchparticipantsCollectinginformationSeekingconsentProvidingincentivesSeekingsensitiveinformationThepossibilityofcausingharmtoparticipantsMaintainingconfidentialityEthicalissuestoconsiderrelatingtotheresearcherAvoidingbiasProvisionordeprivationofatreatmentUsinginappropriateresearchmethodologyIncorrectreportingInappropriateuseoftheinformationEthicalissuesregardingthesponsoringorganisationRestrictionsimposedbythesponsoringorganisationThemisuseofinformation
SummarySTEPVIIPROCESSINGANDDISPLAYINGDATA
15ProcessingdataPartone:DataprocessinginquantitativestudiesEditingCodingParttwo:DataprocessinginqualitativestudiesContentanalysisinqualitativeresearch–anexampleTheroleofstatisticsinresearchSummary
16DisplayingdataMethodsofcommunicatinganddisplayinganalyseddataTextTablesGraphsSummarySTEPVIIIWRITINGARESEARCHREPORT
17WritingaresearchreportWritingaresearchreportDevelopinganoutlineWritingaboutavariableReferencingWritingabibliographySummary
18ResearchmethodologyandpracticeevaluationWhatisevaluation?Whyevaluation?Intervention–development–evaluationprocessPerspectivesintheclassificationofevaluationstudiesTypesofevaluationfromafocusperspectiveEvaluationforprogramme/interventionplanningProcess/monitoringevaluationEvaluatingparticipationofthetargetpopulationEvaluatingservicedeliverymannerImpact/outcomeevaluationCost–benefit/cost-effectivenessevaluationTypesofevaluationfromaphilosophicalperspectiveGoal-centred/objective-orientedevaluationConsumer-oriented/client-centredevaluationImprovement-orientedevaluationHolistic/illuminativeevaluation
Undertakinganevaluation:theprocessStep1:DeterminingthepurposeofevaluationStep2:DevelopingobjectivesorevaluationquestionsStep3:ConvertingconceptsintoindicatorsintovariablesStep4:DevelopingevaluationmethodologyStep5:CollectingdataStep6:AnalysingdataStep7:WritinganevaluationreportStep8:SharingfindingswithstakeholdersInvolvingstakeholdersinevaluationEthicsinevaluationSummary
Appendix:Developingaresearchproject:asetofexercisesforbeginnersGlossaryBibliographyIndex
Figures
1.1Theapplicationsofresearch1.2Typesofresearch
2.1Theresearchjourney2.2Theresearchprocess2.3Thechaptersinthebookinrelationtotheoperationalsteps
3.1aDevelopingatheoreticalframework–therelationshipbetweenmortalityandfertility3.1bTheoreticalframeworkforthestudy‘communityresponsivenessinhealth’3.2Sampleofoutlineofaliteraturereview
4.1Dissectingthesubjectareaofdomesticviolenceintosubareas4.2Stepsinformulatingaresearchproblem–alcoholism4.3Formulatingaresearchproblem–therelationshipbetweenfertilityandmortality4.4Narrowingaresearchproblem–health4.5Characteristicsofobjectives
5.1Typesofvariable5.2Typesofvariableinacausalrelationship5.3Independent,dependentandextraneousvariablesinacausalrelationship5.4Setsofvariablesincounsellingandmarriageproblems5.5Independent,dependent,extraneousandinterveningvariables5.6Activeandattributevariables
6.1Theprocessoftestingahypothesis6.2Two-by-twofactorialexperimenttostudytherelationshipbetweenMCH,NSandinfantmortality6.3Typesofhypothesis6.4TypeIandTypeIIerrorsintestingahypothesis
7.1Factorsaffectingtherelationshipbetweenacounsellingserviceandtheextentofmaritalproblems7.2Therelationshipbetweenteachingmodelsandcomprehension7.3Theproportionattributabletothethreecomponentsmayvarymarkedly7.4Buildingintothedesign
8.1Typesofstudydesign8.2Before-and-after(pre-test/post-test)studydesign8.3Theregressioneffect8.4Thelongitudinalstudydesign8.5aRetrospectivestudydesign8.5bProspectivestudydesign8.5cRetrospective-prospectivestudydesign
8.6Experimentalandnon-experimentalstudies8.7Randomisationinexperiments8.8Theafter-onlydesign8.9Measurementofchangethroughabefore-and-afterdesign8.10Thecontrolexperimentaldesign8.11Double-controldesigns8.12Comparativeexperimentaldesign8.13Theplacebodesign8.14Thecross-overexperimentaldesign8.15Thereplicatedcross-sectionaldesign8.16Actionresearchdesign
9.1Methodsofdatacollection9.2Athree-directionalratingscale9.3Typesofinterview9.4Example1:Wheretogo?Astudyofoccupationalmobilityamongimmigrants9.5Example2:Occupationalredeployment–astudyofoccupationalredeploymentamongstate
governmentemployees9.6Examplesofclosedquestions9.7Examplesofopen-endedquestions
10.1Anexampleofacategoricalscale10.2Anexampleofaseven-pointnumericalscale10.3Anexampleofascalewithstatementsreflectingvaryingdegreesofanattitude10.4TheprocedureforconstructingaLikertscale10.5Scoringpositiveandnegativestatements10.6Calculatinganattitudinalscore10.7TheprocedureforconstructingtheThurstonescale
12.1Theconceptofsampling12.2Typesofsamplinginquantitativeresearch12.3Theprocedureforusingatableofrandomnumbers12.4Theprocedureforselectingasimplerandomsample12.5Theprocedureforselectingastratifiedsample12.6Theconceptofclustersampling12.7Snowballsampling12.8Theprocedureforselectingasystematicsample12.9Systematicsampling
15.1Stepsindataprocessing15.2Exampleofquestionsfromasurvey15.3Someselectedresponsestotheopen-endedquestioninFigure15.215.4Somequestionsfromasurvey–respondent315.5Somequestionsfromasurvey–respondent5915.6Somequestionsfromasurvey–respondent8115.7Anexampleofcodeddataonacodesheet15.8Manualanalysisusinggraphpaper
16.1Thestructureofatable16.2aTwo-dimensionalhistogram16.2bThree-dimensionalhistogram16.2cTwo-dimensionalhistogramwithtwovariables16.3Barcharts16.4Thestackedbarchart16.5The100percentbarchart16.6Thefrequencypolygon16.7Thecumulativefrequencypolygon16.8Thestem-and-leafdisplay16.9Two-andthree-dimensionalpiecharts16.10Thelinediagramortrendcurve16.11Theareachart16.12Thescattergram
18.1Theconceptofevaluation18.2Theintervention–development–evaluationmodel18.3Perspectivesintheclassificationofevaluationstudies18.4Aspectsofprocessevaluation18.5Reflexivecontroldesign18.6Interruptedtime-seriesdesign18.7Replicatedcross-sectionaldesign18.8Convertingconceptsintoindicatorsintovariables18.9Anexampleofconvertingconceptsintoquestions
Tables
1.1Typesofresearchstudiesfromtheperspectiveofobjectives
2.1Differencesbetweenqualitativeandquantitativeresearch
3.1Somecommonlyusedelectronicdatabasesinpublichealth,sociology,educationandbusinessstudies
4.1Aspectsofaresearchproblem4.2Operationalisationofconceptsandthestudypopulations
5.1Examplesofconceptsandvariables5.2Convertingconceptsintovariables5.3Categorical/continuousandquantitative/qualitativevariables5.4Characteristicsandexamplesofthefourmeasurementscales
9.1Guidelinesforconstructingaresearchinstrument
10.1Therelationshipbetweenattitudinalandmeasurementscales
12.1Thedifferencebetweensamplestatisticsandthepopulationmean12.2Thedifferencebetweenasampleandapopulationaverage12.3Selectingasampleusingatableforrandomnumbers12.4Selectedelementsusingthetableofrandomnumbers
13.1Developingatime-frameforyourstudy
15.1Anexampleofacodebook
16.1Respondentsbyage(frequencytableforonepopulation)16.2Respondentsbyage(frequencytablecomparingtwopopulations)16.3Respondentsbyattitudetowardsuraniumminingandage(cross-tabulation)16.4Attitudetowardsuraniumminingbyageandgender16.5Ageandincomedata
18.1Typesofevaluationfromtheperspectiveofitsfocusandthequestionstheyaredesignedtoanswer
Preface
This book is based upon my experiences in research as a student, practitioner and teacher. ThedifficultiesIfacedinunderstandingresearchasastudent,mydiscoveriesaboutwhatwasapplicableandinapplicableinthefieldasapractitioner,andmydevelopmentoftheabilityeffectivelytocommunicatedifficult concepts in simple languagewithout sacrificing technicality and accuracy as a teacher havebecomethebasisofthisbook.Researchmethodologyistaughtasasupportingsubjectinseveralwaysinmanyacademicdisciplines
suchashealth,education,psychology,socialwork,nursing,publichealth,librarystudiesandmarketingresearch. The core philosophical base for this book comes frommy conviction that, although thesedisciplinesvaryincontent,theirbroadapproachtoaresearchenquiryissimilar.Thisbook,therefore,isaddressedtotheseacademicdisciplines.It is true that some disciplines place greater emphasis on quantitative research and some on
qualitativeresearch.Myownapproachtoresearchisacombinationofboth.Firstly,itistheobjectivethat should decidewhether a study be carried out adopting a qualitative or a quantitative approach.Secondly, in real life most research is a combination of both methods. Though they differ in thephilosophy thatunderpins theirmodeof enquiry, to agreat extent theirbroadapproach toenquiry issimilar.Thequantitative researchprocess is reasonablywell structuredwhereas thequalitativeone isfairlyunstructured,andthesearetheirrespectivestrengthsaswellasweaknesses.Istronglybelievethatbothareimportanttoportrayacompletepicture.Inaddition,thereareaspectsofquantitativeresearchthat are qualitative in nature. It depends upon how a piece of information has been collected andanalysed.ThereforeIfeelverystronglythatagoodresearcherneedstohavebothtypesofskill.Ifollowaqualitative–quantitative–qualitativeapproachtoanenquiry.Thisbook,therefore,hasbeenwrittentoprovidetheoreticalinformationinanoperationalmanneraboutmethods,proceduresandtechniquesthatareusedinbothapproaches.Researchasasubjectistaughtatdifferentlevels.Thebookisdesignedspecificallyforstudentswho
arenewcomerstoresearchandwhomayhaveapsychologicalbarrierwithregardtothesubject.Ihavetherefore not assumed any previous knowledge on the part of the reader; I have omitted detaileddiscussion of aspects that may be inappropriate for beginners; I have used many flow charts andexamples to communicate concepts; and areas covered in the book follow a ‘simple to complex’approachintermsoftheirdiscussionandcoverage.The structure of this book,which is based on themodel developed duringmy teaching career, is
designedtobepractical.Thetheoreticalknowledgethatconstitutesresearchmethodologyisthereforeorganised around the operational steps that form this research process for both quantitative andqualitative research. All the information needed to take a particular step, during the actual researchjourney,isprovidedinoneplace.Theneededinformationisorganisedinchaptersandeachchapterisdevoted to a particular aspect of that step (see Figure 2.3). For example, ‘Formulating a researchproblem’isthefirstoperationalstepintheresearchprocess.Forformulatinga‘good’researchproblem,inmyopinion,youneedtoknowhowtoreviewtheliterature,formulatearesearchproblem,dealwithvariables and their measurement, and construct hypotheses. Hence, under this step, there are fourchapters. The information they providewill enable you to formulate a problem that is researchable.These chapters are titled: ‘Reviewing the literature’, ‘Formulating a research problem’, ‘Identifying
variables’and‘Constructinghypotheses’.Similarly,fortheoperationalstep,stepIII,‘Constructinganinstrument fordatacollection’, thechapters titled‘Selectingamethodofdatacollection’, ‘Collectingdatausingattitudinalscales’and‘Establishingthevalidityandreliabilityofaresearchinstrument’willprovidesufficientinformationforyoutodevelopaninstrumentfordatacollectionforyourstudy.Foreveryaspectateachstep,asmorgasbordofmethods,models,techniquesandproceduresisprovidedforboth quantitative and qualitative studies in order for you to build your knowledge base in researchmethodology and also to help you to select the most appropriate ones when undertaking your ownresearch.Itismybeliefthatasoundknowledgeofresearchmethodologyisessentialforundertakingavalid
study.Toansweryour researchquestions, up toStepV, ‘Writing a researchproposal’, knowledgeofresearchmethodsiscrucialasthisenablesyoutodevelopaconceptualframeworkwhichissoundandhasmeritsforundertakingyourresearchendeavourwithconfidence.Havingcompletedthepreparatorywork,thestepsthatfollowaremorepracticalinnature,thequalityofwhichentirelydependsuponthesoundnessofthemethodologyyouproposedinyourresearchproposal.Statisticsandcomputersplayasignificant role in research but their application is mainly after the data has been collected. Tome,statisticsareusefulinconfirmingorcontradictingconclusionsdrawnfromsimplylookingatanalyseddata, in providing an indication of themagnitude of the relationship between two ormore variablesunder study, in helping to establish causality, and in ascertaining the level of confidence that can beplaced in your findings. A computer’s application is primarily in data analysis, the calculation ofstatistics,wordprocessingandthegraphicpresentationofdata.Itsavestimeandmakesiteasierforyoutoundertaketheseactivities;however,youneedtolearnthisadditionalskill.Thisbookdoesnotincludestatisticsorinformationaboutcomputers.Thethirdeditionofthebookincorporatessomeofthesuggestionsmadebythereviewers,colleagues
andstudentsonthefirstandsecondeditions.Therearesomemajorchangesinthethirdedition:
Ihavetakenaveryboldstepinbreakingdown,wherepossible,thewallbetweenqualitativeandquantitativeresearchbydescribingbothmethodologiesparalleltooneanotherwithinacommonframework.Alotmoreinformationonqualitativeresearchhasbeenaddedandintegratedwiththecurrenteight-stepresearchmodel.Now,almosteachchapterhasanewsectionthatisspecificallydevotedtoinformationrelatedtoqualitativeresearchpertainingtothemainthemeofthechapter.Forexample,Chapter9,‘Selectingamethodofdatacollection’,nowhasasection‘Methodsofdatacollectioninqualitativeresearch’thatspecificallydiscussesthemajormethodsofdatacollectioninqualitativestudies.Similarly,Chapter8,‘Selectingastudydesign’,hasasection‘Studydesignsinqualitativeresearch’thatisdevotedtothedesignsdominantlyusedinqualitativeresearch.Asfaraspossibleeachchapteralsohasinformationonotheraspectsofqualitativeresearchalongwiththeexistingquantitativebodyofknowledge.Morein-depthfieldexamples,baseduponactualexperiences,havebeenincorporatedtoexplainproceduresandmethods.Exercises,apartoftheAppendix,havenowbeenthoroughlyrevisedwiththeexpectationthatthosewhoaredevelopingaresearchprojectcanoperationalisethetheoreticalknowledgeinanactualsituationtoevaluatetheapplicationoftheorytopracticeinadditiontodevelopingtheirresearchproject.Aglossaryoftechnicaltermsisanewadditiontothisedition.Thiswillprovidestudentswithreadilyavailabledefinitionsandmeaningsoftechnicaltermsinoneplace.Titlepagesdividingchaptersandoperationalstepshavenowbeenredesignedtoprovidegreaterclarityaswellasinformingstudentsinadvancewhattheyareexpectedtolearninachapter.Also,
eachchapterhasalistofkeywordsthatstudentsarelikelytoencounterinthechapter.Inplacesthelanguagehasbeenchangedtoenhanceflow,understandingandeaseofreading.
Iamgratefultoanumberofpeoplewhohavehelpedmeinthewritingofthisbook.Firstofall,tomystudents,whohave taughtmehow to teach researchmethods.Thebasic structureof this book is anoutcome of the feedback I have received from them over the years. How, and at what stage of theresearchprocess,aconceptoraprocedureshouldbetaught,Ihavelearntfrommystudents.Ithankfullyacknowledgetheircontributiontothisbook.Iamextremelygratefultoafriendandcolleague,DrNormaWatson,whoseeffortsineditingthefirst
edition were of immense help. The book would not have come to its present stage without herunconditionalhelp.IalsothankProfessorDenisLadbrook,afriendandcolleague,forhiscontinuousencouragementand
support.
RanjitKumar
CHAPTER1Research:AWayofThinking
Inthischapteryouwilllearnabout:
SomeofthereasonsfordoingresearchHowresearchcanbeusedtogatherevidencetoinformyourpracticeTheapplicationsofresearchCharacteristicsandrequirementsoftheresearchprocessTypesofresearchfromtheperspectiveofapplications,objectivesandenquirymodesResearchparadigms
Keywords: applied research, controlled, correlational research, descriptiveresearch, empirical, explanatory research, exploratory research, evidence-basedpractice, interpretive paradigm, positivistic paradigm, pure research, qualitativeresearch, quantitative research, reliability, research, structured and unstructuredenquiries,systematic,validity.
Research:anintegralpartofyourpractice
Researchisundertakenwithinmostprofessions.Morethanasetofskills,researchisawayofthinking:examining critically the various aspects of your day-to-day professional work; understanding andformulating guiding principles that govern a particular procedure; and developing and testing newtheoriesthatcontributetotheadvancementofyourpracticeandprofession.Itisahabitofquestioningwhatyoudo,andasystematicexaminationofclinicalobservationstoexplainandfindanswersforwhatyouperceive,withaviewtoinstitutingappropriatechangesforamoreeffectiveprofessionalservice.Letustakesomedisciplinesasexamples.Supposeyouareworkinginthefieldofhealth.Youmaybeafront-lineserviceprovider,supervisor
orhealthadministrator/planner.Youmaybeinahospitalorworkingasanoutreachcommunityhealthworker. You may be a nurse, doctor, occupational therapist, physiotherapist, social worker or otherparamedic. In any of these positions, some of the following questions may come to your mind orsomeoneelsemayaskyoufortheiranswers:
HowmanypatientsdoIseeeveryday?Whataresomeofthemostcommonconditionsprevalentamongmypatients?Whatarethecausesoftheseconditions?Whydosomepeoplehaveaparticularconditionwhereasothersdonot?Whatarethehealthneedsofthecommunity?Whatarethebenefitsofthisprogrammetothecommunity?HowdoIdemonstratetheeffectivenessofmyservice?Whydosomepeopleusetheservicewhileothersdonot?Whatdopeoplethinkabouttheservice?Howsatisfiedarepatientswiththeservice?Howeffectiveistheservice?Howcantheservicebeimproved?
Youcanaddmanyotherquestionstothislist.Attimesitmaybepossibletoignorethesequestionsbecauseofthelevelatwhichyouwork,atothertimesyoumaymakeanefforttofindanswersonyourowninitiative,or,sometimes,youmayberequired toobtainanswersforeffectiveadministrationandplanning.Letustakeanotherdiscipline:businessstudies.Assumeyouworkintheareaofmarketing.Again,
youcanworkatdifferentlevels:asasalesperson,salesmanagerorsalespromotionexecutive.Thelistofquestionsthatmaycometoyourmindcanbeendless.Thetypesofquestionsandtheneedtofindanswerstothemwillvarywiththelevelatwhichyouworkintheorganisation.Youmayjustwanttofindoutthemonthlyfluctuationinthesalesofaparticularproduct,oryoumaybeaskedtodevelopanR&D strategic plan to compete for a greater share of themarket for the products produced by yourcompany.Thelistofquestionsthatmaycometomindcouldbeendless.Forexample:
Whatisthebeststrategytopromotethesaleofaparticularproduct?HowmanysalespersonsdoIneed?Whatistheeffectofaparticularadvertisingcampaignonthesaleofthisproduct?Howsatisfiedaretheconsumerswiththisproduct?Howmuchareconsumerspreparedtospendonthisproduct?Whatdoconsumerslikeordislikeaboutthisproduct?Whattypeofpackagingdoconsumerspreferforthisproduct?Whattrainingdothesalespersonsneedtopromotethesaleofthisproduct?Whataretheattributesofagoodsalesperson?
To take a different example, let us assume that you work as a psychologist, counsellor or socialworker.Whileengaginginthehelpingprocessyoumayaskyourself(orsomeoneelsemayaskyou)thefollowingquestions:
Whataremyclients’mostcommonpresentingproblems?Whataretheirmostcommonunderlyingproblems?Whatisthesocioeconomicbackgroundofmyclients?WhyamIsuccessfulincertaincasesandnotinothers?Whatresourcesareavailableinthecommunitytohelpaclientwithaparticularneed?Whatinterventionstrategiesareappropriateforthisproblem?Howsatisfiedaremyclientswithmyservices?
As a supervisor, administrator or manager of an agency, again different questions relating toeffectivenessandefficiencyofaservicemaycometoyourmind.Forexample:
Howmanypeoplearecomingtomyagency?Whatarethesocioeconomic–demographiccharacteristicsofmyclients?Howmanycasesinadaycanaworkereffectivelyhandle?Whydosomepeopleusetheservicewhileothersdonot?Howeffectiveistheservice?Whatarethemostcommonneedsofclientswhocometothisagency?Whatarethestrengthsandweaknessesoftheservice?Howsatisfiedaretheclientswiththeservice?HowcanIimprovethisserviceformyclients?
Asaprofessionalyoumightbeinterestedinfindinganswerstotheoreticalquestions,suchas:
Whichisthemosteffectiveinterventionforaparticularproblem?WhatcausesXorwhataretheeffectsofY?Whatistherelationshipbetweentwophenomena?HowdoImeasuretheself-esteemofmyclients?HowdoIascertainthevalidityofmyquestionnaire?Whatisthepatternofprogrammeadoptioninthecommunity?Whichisthebestwayoffindingoutcommunityattitudestowardsanissue?Whichisthebestwaytofindouttheeffectivenessofaparticulartreatment?HowcanIselectanunbiasedsample?Whatisthebestwaytofindoutaboutthelevelofmarriagesatisfactionamongmyclients?
Inthisageofconsumerismyoucannotaffordtoignoretheconsumersofaservice.Consumershavetherighttoaskquestionsaboutthequalityandeffectivenessoftheservicetheyarereceivingandyou,as the service provider, have an obligation to answer their questions. Some of the questions that aconsumermayaskare:
HoweffectiveistheservicethatIamreceiving?AmIgettingvalueformoney?Howwelltrainedaretheserviceproviders?
Mostprofessionsthatareinthehumanserviceindustrywouldlendthemselvestothequestionsraisedaboveandyouasa serviceprovider shouldbewellprepared toanswer them.Research isoneof thewaystohelpyouanswersuchquestionsobjectively.
Research:awaytogatherevidenceforyourpractice
Evidence-basedpractice (EBP) is thedeliveryof servicesbasedupon researchevidenceabout theireffectiveness; theserviceprovider’sclinical judgementastothesuitabilityandappropriatenessoftheservice for a client; and the client’s own preference as to the acceptance of the service. EBP is fastbecominga servicedeliverynormamongmanyprofessions.Though itsorigin is credited tomedical
practice,EBPhasbecomean importantpartofmanyotherprofessions suchasnursing, alliedhealthservices, mental health, community health, social work, psychology and teaching. It is now beingpromotedasanacceptableandscientificmethodforpolicyformulationandpracticeassessment.TheconceptofEBPencouragesprofessionalsandotherdecision-makers touseevidenceregarding
theeffectivenessofaninterventioninconjunctionwiththecharacteristicsandcircumstanceofaclientand their own professional judgement to determine the appropriateness of an intervention whenprovidingaservicetoaclient.Inthisageofaccountability,youasaprofessionalmustbeaccountabletoyourclientsaswellasyourprofession.Itisasapartofthisaccountabilitythatyouneedtodemonstratetheeffectivenessoftheservice(s)youprovide.Research is one of the ways of collecting accurate, sound and reliable information about the
effectivenessofyourinterventions,therebyprovidingyouwithevidenceofitseffectiveness.Asserviceprovidersandprofessionals,weusetechniquesandproceduresdevelopedbyresearchmethodologiststoconsolidate, improve,develop, refineandadvanceclinicalaspectsofourpractice toserveourclientsbetter.
Applicationsofresearch
Very little research in the field is pure in nature. That is, very few people do research in researchmethodologyperse.Mostresearchisappliedresearch,whichhaswideapplicationinmanydisciplines.Everyprofessionusesresearchmethodsinvaryingamountsinmanyareas.Theyusethemethodsandprocedures developed by research methodologists in order to increase understanding in their ownprofession and to advance the professional knowledge base. It is through the application of researchmethodologythattheystrengthenandadvancetheirownprofession.Examineyourownfield.Youwillfindthatitsprofessionalpracticefollowsproceduresandpracticestestedanddevelopedbyothersoveralongperiodoftime.Itisinthistestingprocessthatyouneedresearchskills,thedevelopmentsofwhichfallinthecategoryofpureresearch.Asamatteroffact,thevalidityofyourfindingsentirelydependsuponthesoundnessoftheresearchmethodsandproceduresadoptedbyyou.Within anyprofession,whereyoudirectly or indirectly provide a service, such as health (nursing,
occupationaltherapy,physiotherapy,communityhealth,healthpromotionandpublichealth),education,psychologyorsocialwork,theapplicationofresearchcanbeviewedfromfourdifferentperspectives:
1. theserviceprovider;2. theserviceadministrator,managerand/orplanner;3. theserviceconsumer;and4. theprofessional.
TheseperspectivesaresummarisedinFigure1.1.Thoughitisimpossibletolistalltheissuesineverydiscipline, this frameworkcanbeapplied tomostdisciplinesandsituations in thehumanitiesand thesocialsciences.Youshouldbeabletousethistoidentify,fromtheviewpointoftheaboveperspectives,thepossibleissuesinyourownacademicfieldwhereresearchtechniquescanbeusedtofindanswers.
Research:whatdoesitmean?
Thereareseveralwaysofobtaininganswerstoyourprofessionalquestions.Thesemethodsrangefrom
the fairly informal, based upon clinical impressions, to the strictly scientific, adhering to theconventionalexpectationsofscientificprocedures.Researchisoneofthewaystofindanswerstoyourquestions.Whenyousaythatyouareundertakingaresearchstudytofindoutanswers toaquestion,youareimplyingthattheprocessbeingapplied:
1. isbeingundertakenwithinaframeworkofasetofphilosophies;2. usesprocedures,methodsandtechniquesthathavebeentestedfortheirvalidityandreliability;3. isdesignedtobeunbiasedandobjective.
Your philosophical orientation may stem from one of the several paradigms and approaches inresearch – positivist, interpretive, phenomenology, action or participatory, feminist, qualitative,quantitative–andtheacademicdisciplineinwhichyouhavebeentrained.Theconceptof‘validity’canbeappliedtoanyaspectoftheresearchprocess.Itensuresthatinaresearchstudycorrectprocedureshave been applied to find answers to a question. ‘Reliability’ refers to the quality of ameasurementprocedurethatprovidesrepeatabilityandaccuracy.‘Unbiasedandobjective’meansthatyouhavetakeneach step in an unbiasedmanner and drawn each conclusion to the best of your ability andwithoutintroducingyourownvested interest.The authormakes adistinctionbetweenbias and subjectivity.Subjectivity is an integral part of your way of thinking that is ‘conditioned’ by your educationalbackground,discipline,philosophy,experienceandskills.Forexample, apsychologistmay lookat apieceofinformationdifferentlyfromthewayinwhichananthropologistorahistorianlooksatit.Bias,on theotherhand, is a deliberate attempt to either conceal or highlight something.Adherence to thethreecriteriamentionedaboveenablestheprocesstobecalled‘research’.Therefore,whenyousayyouareundertakingaresearchstudytofindtheanswertoaquestion,thisimpliesthatthemethod(s)youareadoptingfulfilstheseexpectations(discussedlaterinthechapter).
FIGURE1.1Theapplicationsofresearch
However, the degree to which these criteria are expected to be fulfilled varies from discipline todiscipline and so the meaning of ‘research’ differs from one academic discipline to another. Forexample,theexpectationsoftheresearchprocessaremarkedlydifferentbetweenthephysicalandthesocial sciences. In the physical sciences a research endeavour is expected to be strictly controlled ateachstep,whereas in the social sciences rigidcontrolcannotbeenforcedandsometimes isnotevendemanded.Within thesocialsciences the levelofcontrol requiredalsovariesmarkedlyfromonediscipline to
another,associalscientistsdifferovertheneedfortheresearchprocesstomeettheaboveexpectations.Despite these differences amongdisciplines, their broad approach to enquiry is similar.The researchmodel,thebasisofthisbook,isbaseduponthisbroadapproach.Asbeginnersinresearchyoushouldunderstandthatresearchisnotalltechnical,complex,statistics
andcomputers. Itcanbeaverysimpleactivitydesignedtoprovideanswers toverysimplequestionsrelating to day-to-day activities. On the other hand, research procedures can also be employed toformulate intricate theories or laws that govern our lives. The difference between research and non-researchactivityis,asmentioned,inthewaywefindanswerstoourresearchquestions.Foraprocesstobecalledresearch,itisimportantthatitmeetscertainrequirementsandpossessescertaincharacteristics.Toidentifytheserequirementsandcharacteristicsletusexaminesomedefinitionsofresearch:
Thewordresearchiscomposedoftwosyllables,reandsearch.Thedictionarydefinestheformer
asaprefixmeaningagain,aneworoveragainandthelatterasaverbmeaningtoexaminecloselyand carefully, to test and try, or to probe. Together they form a noun describing a careful,systematic, patient study and investigation in some field of knowledge, undertaken to establishfactsorprinciples.(Grinnell1993:4)
Grinnellfurtheradds:‘researchisastructuredinquirythatutilisesacceptablescientificmethodologytosolveproblemsandcreatesnewknowledgethatisgenerallyapplicable.’(1993:4)Lundberg(1942)drawsaparallelbetweenthesocialresearchprocess,whichisconsideredscientific,
andtheprocessthatweuseinourdailylives.Accordingtohim:
Scientificmethodsconsistofsystematicobservation,classificationandinterpretationofdata.Now,obviously,thisprocessisoneinwhichnearlyallpeopleengageinthecourseoftheirdailylives.The main difference between our day-to-day generalisations and the conclusions usuallyrecognised as scientific method lies in the degree of formality, rigorousness, verifiability andgeneralvalidityofthelatter.(Lundberg1942:5)
Burns(1997:2)definesresearchas‘asystematicinvestigationtofindanswerstoaproblem’.According to Kerlinger (1986: 10), ‘scientific research is a systematic, controlled empirical and
critical investigation of propositions about the presumed relationships about various phenomena’.Bulmer (1977: 5) states: ‘Nevertheless sociological research, as research, is primarily committed toestablishingsystematic,reliableandvalidknowledgeaboutthesocialworld.’
Theresearchprocess:characteristicsandrequirements
From these definitions it is clear that research is a process for collecting, analysing and interpretinginformation to answer questions. But to qualify as research, the process must have certaincharacteristics: it must, as far as possible, be controlled, rigorous, systematic, valid and verifiable,empiricalandcritical.Letusbrieflyexaminethesecharacteristicstounderstandwhattheymean:
Controlled–Inreallifetherearemanyfactorsthataffectanoutcome.Aparticulareventisseldomtheresultofaone-to-onerelationship.Somerelationshipsaremorecomplexthanothers.Mostoutcomesareasequeltotheinterplayofamultiplicityofrelationshipsandinteractingfactors.Inastudyofcause-and-effectrelationshipsitisimportanttobeabletolinktheeffect(s)withthecause(s)andviceversa.Inthestudyofcausation,theestablishmentofthislinkageisessential;however,inpractice,particularlyinthesocialsciences,itisextremelydifficult–andoftenimpossible–tomakethelink.
The concept of control implies that, in exploring causality in relation to twovariables, you set upyourstudy inaway thatminimises theeffectsofother factorsaffecting the relationship.Thiscanbeachieved to a large extent in the physical sciences, asmost of the research is done in a laboratory.However, inthesocialsciencesit isextremelydifficultasresearchiscarriedoutonissuesrelatingtohumanbeingslivinginsociety,wheresuchcontrolsareimpossible.Therefore,inthesocialsciences,asyoucannotcontrolexternalfactors,youattempttoquantifytheirimpact.
Rigorous–Youmustbescrupulousinensuringthattheproceduresfollowedtofindanswerstoquestionsarerelevant,appropriateandjustified.Again,thedegreeofrigourvariesmarkedlybetweenthephysicalandthesocialsciencesandwithinthesocialsciences.Systematic–Thisimpliesthattheproceduresadoptedtoundertakeaninvestigationfollowacertainlogicalsequence.Thedifferentstepscannotbetakeninahaphazardway.Someproceduresmustfollowothers.Validandverifiable–Thisconceptimpliesthatwhateveryouconcludeonthebasisofyourfindingsiscorrectandcanbeverifiedbyyouandothers.Empirical–Thismeansthatanyconclusionsdrawnarebaseduponhardevidencegatheredfrominformationcollectedfromreal-lifeexperiencesorobservations.Critical–Criticalscrutinyoftheproceduresusedandthemethodsemployediscrucialtoaresearchenquiry.Theprocessofinvestigationmustbefoolproofandfreefromanydrawbacks.Theprocessadoptedandtheproceduresusedmustbeabletowithstandcriticalscrutiny.
Foraprocesstobecalledresearch,itisimperativethatithastheabovecharacteristics.
FIGURE1.2Typesofresearch
Typesofresearch
Typesofresearchcanbelookedatfromthreedifferentperspectives(Figure1.2):
1. applicationsofthefindingsoftheresearchstudy;2. objectivesofthestudy;3. modeofenquiryusedinconductingthestudy.
Theclassificationofthetypesofastudyonthebasisoftheseperspectivesisnotmutuallyexclusive:that is, a researchstudyclassified from theviewpointof ‘application’canalsobeclassified from theperspectives of ‘objectives’ and ‘enquiry mode’ employed. For example, a research project may beclassifiedaspureorappliedresearch(fromtheperspectiveofapplication),asdescriptive,correlational,explanatoryorexploratory(fromtheperspectiveofobjectives)andasqualitativeorquantitative(fromtheperspectiveoftheenquirymodeemployed).
Typesofresearch:applicationperspective
If you examine a research endeavour from the perspective of its application, there are two broadcategories:pureresearchandappliedresearch.Inthesocialsciences,accordingtoBailey(1978:17):
Pure research involves developing and testing theories and hypotheses that are intellectuallychallengingtotheresearcherbutmayormaynothavepracticalapplicationatthepresenttimeorinthefuture.Thussuchworkofteninvolvesthetestingofhypothesescontainingveryabstractandspecialisedconcepts.
Pure research is also concerned with the development, examination, verification and refinement ofresearch methods, procedures, techniques and tools that form the body of research methodology.Examplesofpureresearchincludedevelopingasamplingtechniquethatcanbeappliedtoaparticularsituation; developing amethodology to assess the validity of a procedure; developing an instrument,say,tomeasurethestresslevelinpeople;andfindingthebestwayofmeasuringpeople’sattitudes.Theknowledgeproducedthroughpureresearchissoughtinordertoaddtotheexistingbodyofknowledgeofresearchmethods.Most of the research in the social sciences is applied. In other words, the research techniques,
procedures andmethods that form thebodyof researchmethodologyare applied to thecollectionofinformationaboutvariousaspectsofasituation,issue,problemorphenomenonsothattheinformationgathered can be used in other ways – such as for policy formulation, administration and theenhancementofunderstandingofaphenomenon.
Typesofresearch:objectivesperspective
Ifyouexaminearesearchstudyfromtheperspectiveofitsobjectives,broadlyaresearchendeavourcanbeclassifiedasdescriptive,correlational,explanatoryorexploratory.A research study classified as adescriptive study attempts to describe systematically a situation,
problem,phenomenon,serviceorprogramme,orprovidesinformationabout,say,thelivingconditionsof a community,ordescribesattitudes towardsan issue.For example, itmayattempt todescribe thetypesofserviceprovidedbyanorganisation,theadministrativestructureofanorganisation,thelivingconditionsofAboriginalpeopleintheoutback,theneedsofacommunity,whatitmeanstogothrougha divorce, how a child feels living in a housewith domestic violence, or the attitudes of employeestowardsmanagement.Themainpurposeofsuchstudiesistodescribewhatisprevalentwithrespecttotheissue/problemunderstudy.The main emphasis in a correlational study is to discover or establish the existence of a
relationship/association/interdependencebetweentwoormoreaspectsofasituation.Whatistheimpactofanadvertisingcampaignonthesaleofaproduct?Whatistherelationshipbetweenstressfullivingandtheincidenceofheartattack?Whatistherelationshipbetweenfertilityandmortality?Whatistherelationship between technology and unemployment? What is the effect of a health service on thecontrol of a disease, or the home environment on educational achievement? These studies examinewhether there is a relationship between two or more aspects of a situation or phenomenon and,therefore,arecalledcorrelationalstudies.Explanatoryresearchattemptstoclarifywhyandhowthereisarelationshipbetweentwoaspectsof
asituationorphenomenon.Thistypeofresearchattemptstoexplain,forexample,whystressfullivingresultsinheartattacks;whyadeclineinmortalityisfollowedbyafertilitydecline;orhowthehome
environmentaffectschildren’slevelofacademicachievement.The fourth typeof research, from theviewpointof theobjectivesofa study, iscalledexploratory
research.Thisiswhenastudyisundertakenwiththeobjectiveeithertoexploreanareawherelittleisknown or to investigate the possibilities of undertaking a particular research study.When a study iscarriedouttodetermineitsfeasibilityitisalsocalledafeasibilitystudyorapilotstudy.It isusuallycarriedoutwhena researcherwants toexploreareasaboutwhich s/hehas littleornoknowledge.Asmall-scalestudyisundertakentodecideifitisworthcarryingoutadetailedinvestigation.Onthebasisoftheassessmentmadeduringtheexploratorystudy,afullstudymayeventuate.Exploratorystudiesarealsoconducted todevelop, refineand/or testmeasurement toolsandprocedures.Table1.1 shows thetypesofresearchstudyfromtheviewpointofobjectives.Although,theoretically,aresearchstudycanbeclassifiedinoneoftheaboveobjectives–perspective
categories,inpractice,moststudiesareacombinationofthefirstthree;thatis,theycontainelementsofdescriptive,correlationalandexplanatoryresearch.Inthisbooktheguidelinessuggestedforwritingaresearchreportencourageyoutointegratetheseaspects.
Typesofresearch:modeofenquiryperspective
Thethirdperspectiveinourtypologyofresearchconcernstheprocessyouadopttofindanswerstoyourresearchquestions.Broadly,therearetwoapproachestoenquiry:
1. thestructuredapproach;2. theunstructuredapproach.
Inthestructuredapproacheverythingthatformstheresearchprocess–objectives,design,sample,andthe questions that you plan to ask of respondents – is predetermined.The unstructured approach, bycontrast, allows flexibility in all these aspects of the process. The structured approach is moreappropriate to determine the extent of a problem, issue or phenomenon, whereas the unstructuredapproach is predominantly used to explore itsnature, in other words, variation/diversity per se in aphenomenon, issue, problem or attitude towards an issue. For example, if you want to research thedifferent perspectives of an issue, the problems experienced by people living in a community or thedifferent views people hold towards an issue, then these are better explored using unstructuredenquiries.On theother hand, to findout howmanypeople have a particular perspective, howmanypeople have a particular problem, or howmany people hold a particular view, you need to have astructured approach to enquiry. Before undertaking a structured enquiry, in the author’s opinion, anunstructuredenquirymustbeundertakentoascertainthediversityinaphenomenonwhichcanthenbequantifiedthroughthestructuredenquiry.Bothapproacheshavetheirplaceinresearch.Bothhavetheirstrengths and weaknesses. Therefore, you should not ‘lock’ yourself solely into a structured orunstructuredapproach.
TABLE1.1Typesofresearchstudiesfromtheperspectiveofobjectives
Thestructuredapproachtoenquiry isusuallyclassifiedasquantitativeresearchandunstructuredasqualitativeresearch.Other distinctions betweenquantitative and qualitative research are outlined inTable2.1 inChapter2.The choice betweenquantitative and qualitative approaches (or structured orunstructured)shoulddependupon:
Aimofyourenquiry–exploration,confirmationorquantification.Useofthefindings–policyformulationorprocessunderstanding.
Thedistinctionbetweenquantitativeandqualitativeresearch,inadditiontothestructured/unstructuredprocess of enquiry, is also dependent upon some other considerationswhich are briefly presented inTable2.1.A study is classified as qualitative if the purpose of the study is primarily to describe a situation,
phenomenon,problemorevent;iftheinformationisgatheredthroughtheuseofvariablesmeasuredonnominalorordinalscales(qualitativemeasurementscales);andiftheanalysisisdonetoestablishthevariation in the situation, phenomenon or problem without quantifying it. The description of anobserved situation, the historical enumeration of events, an account of the different opinions peoplehave about an issue, and a description of the living conditions of a community are examples ofqualitativeresearch.On theotherhand, the study isclassifiedasquantitative ifyouwant toquantify thevariation in a
phenomenon, situation, problemor issue; if information is gatheredusingpredominantlyquantitative
variables; and if the analysis is geared to ascertain the magnitude of the variation. Examples ofquantitativeaspectsofaresearchstudyare:Howmanypeoplehaveaparticularproblem?Howmanypeopleholdaparticularattitude?Theuseofstatisticsisnotanintegralpartofaquantitativestudy.Themainfunctionofstatisticsisto
act as a test to confirm or contradict the conclusions that you have drawn on the basis of yourunderstandingofanalyseddata.Statistics,amongotherthings,helpyoutoquantifythemagnitudeofanassociationorrelationship,provideanindicationoftheconfidenceyoucanplaceinyourfindingsandhelpyoutoisolatetheeffectofdifferentvariables.Itisstronglyrecommendedthatyoudonot‘lockyourself’intobecomingeithersolelyaquantitative
orsolelyaqualitativeresearcher.Itistruethattherearedisciplinesthatlendthemselvespredominantlyeithertoqualitativeortoquantitativeresearch.Forexample,suchdisciplinesasanthropology,historyand sociology are more inclined towards qualitative research, whereas psychology, epidemiology,education, economics, public health and marketing are more inclined towards quantitative research.However,thisdoesnotmeanthataneconomistorapsychologistneverusesthequalitativeapproach,orthat an anthropologist never uses quantitative information. There is increasing recognition by mostdisciplinesinthesocialsciencesthatbothtypesofresearchareimportantforagoodresearchstudy.Theresearch problem itself should determine whether the study is carried out using quantitative orqualitativemethodologies.Asbothqualitativeandquantitativeapproacheshavetheirstrengthsandweaknesses,andadvantages
anddisadvantages, ‘neitherone ismarkedlysuperior to theother inall respects’ (Ackroyd&Hughes1992: 30). Themeasurement and analysis of the variables aboutwhich information is obtained in aresearchstudyaredependentuponthepurposeofthestudy.Inmanystudiesyouneedtocombinebothqualitativeandquantitativeapproaches.Forexample,supposeyouwanttofindoutthetypesofserviceavailabletovictimsofdomesticviolenceinacityandtheextentoftheirutilisation.Typesofserviceisthe qualitative aspect of the study as finding out about them entails description of the services. Theextent of utilisation of the services is the quantitative aspect as it involves estimating the number ofpeoplewhousetheservicesandcalculatingotherindicatorsthatreflecttheextentofutilisation.
Paradigmsofresearch
Thereare twomainparadigms that form thebasisof research in thesocial sciences. It isbeyond thescopeofthisbooktogointoanydetailaboutthese.Thecrucialquestionthatdividesthetwoiswhetherthe methodology of the physical sciences can be applied to the study of social phenomena. Theparadigmthatisrootedinthephysicalsciencesiscalledthesystematic,scientificorpositivistapproach.Theoppositeparadigmhascometobeknownasthequalitative,ethnographic,ecologicalornaturalisticapproach. The advocates of the two opposing sides have developed their own values, terminology,methodsandtechniquestounderstandsocialphenomena.However,sincethemid-1960stherehasbeena growing recognition that both paradigms have their place. It is the purpose for which a researchactivity is undertaken that should determine the mode of enquiry, hence the paradigm. Toindiscriminatelyapplyoneapproachtoalltheresearchproblemscanbemisleadingandinappropriate.Apositivistparadigmlendsitself tobothquantitativeandqualitativeresearch.However, theauthor
makesadistinctionbetweenqualitativedataontheonehandandqualitativeresearchontheotherasthefirstisconfinedtothemeasurementofvariablesandthesecondtoauseofmethodology.Theauthorbelievesthatnomatterwhatparadigmtheresearcherworkswithin,s/heshouldadhereto
certain values regarding the control of bias, and themaintenance of objectivity in terms of both theresearchprocessitselfandtheconclusionsdrawn.Itistheapplicationofthesevaluestotheprocessof
informationgathering,analysisandinterpretationthatenablesittobecalledaresearchprocess.
SummaryThereareseveralwaysofcollectingandunderstandinginformationandfindinganswerstoyourquestions–researchisoneway.Thedifferencebetweenresearchandotherwaysofobtaininganswerstoyourquestionsisthatinaprocessthatisclassifiedasresearch,youworkwithinaframeworkofasetofphilosophies,usemethodsthathavebeentestedforvalidityandreliability,andattempttobeunbiasedandobjective.Researchhasmany applications.Youneed to have research skills to be an effective service provider, administrator/manager or
planner.Asaprofessionalwhohasaresponsibilitytoenhanceprofessionalknowledge,researchskillsareessential.Thetypologyofresearchcanbelookedatfromthreeperspectives:application,objectivesandtheenquiryprocess.Fromthepoint
ofviewoftheapplicationofresearch, thereisappliedandpureresearch.Mostoftheresearchundertakeninthesocialsciencesisapplied,thefindingsbeingdesignedeitherforuseinunderstandingaphenomenon/issueortobringchangeinaprogramme/situation.Pure research is academic in nature and is undertaken in order to gain knowledge about phenomena that may or may not haveapplicationsinthenearfuture,andtodevelopnewtechniquesandproceduresthatformthebodyofresearchmethodology.Aresearchstudy can be carried out with four objectives: to describe a situation, phenomenon, problem or issue (descriptive research); toestablishorexplorearelationshipbetweentwoormorevariables(correlationalresearch);toexplainwhycertainthingshappenthewaytheydo(explanatoryresearch);andtoexaminethefeasibilityofconductingastudyorexploringasubjectareawherenothingorlittle isknown(exploratoryresearch).Fromthepointofviewofthemodeofenquiry, therearetwotypesofresearch:quantitative(structuredapproach)andqualitative(unstructuredapproach).Themainobjectiveofaqualitativestudyistodescribethevariationanddiversityinaphenomenon,situationorattitudewithaveryflexibleapproachsoastoidentifyasmuchvariationanddiversityaspossible,whereasquantitativeresearch,inaddition,helpsyoutoquantifythevariationanddiversity.Therearemanywhostronglyadvocateacombinedapproachtosocialenquiries.These are the two paradigms that form the basis of social science research. Though these may provide values, terminology,
methodsandtechniquesforyoutoapplytoyourresearch,itisthepurposeofresearchratherthantheparadigmthatshoulddeterminethemodeofenquiry.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Considerhowyouwouldgoaboutconvincingaserviceproviderthatevidence-basedresearchmightbenefitthem.Identifytwoorthreeresearchquestions,relatedtoyourownacademicfieldorprofessionalarea,thatcouldbeansweredbyundertakingeachofthefollowingtypesofresearch:
descriptiveresearch;correlationalresearch;explanatoryresearch;exploratoryresearch.
Considerhowbothunstructuredandstructuredapproachestoresearchcouldbeappliedtoimprovepracticeinyourownprofessionalarea.Criticallyexamineyourownresearchphilosophyinrelationtothetworesearchparadigms.
CHAPTER2TheResearchProcess:AQuickGlance
Inthischapteryouwilllearnabout:
Theeight-stepmodelforcarryingoutresearch
PHASEIDECIDINGWHATTORESEARCH
StepIFormulatingaresearchproblem
PHASEIIPLANNINGARESEARCHSTUDY
StepIIConceptualisingaresearchdesignStepIIIConstructinganinstrumentfordatacollectionStepIVSelectingasampleStepVWritingaresearchproposal
PHASEIIICONDUCTINGARESEARCHSTUDY
StepVICollectingdataStepVIIProcessinganddisplayingdataStepVIIIWritingaresearchreport
Keywords: data,datadisplay,dataprocessing,empiricism,hypotheses, interviewschedule, non-probability sample, primary data, probability sample, qualitativeresearch,questionnaire,rationalism,reliability,researchdesign,researchinstrument,research objectives, research problem, research proposal, sample, sample size,sampling design, secondary data, study design, unstructured interview, validity,variables.
Butmuchadvantagewilloccurifmenofsciencebecometheirownepistemologists,andshowto
theworldbycriticalexpositioninnon-technicaltermstheresultsandmethodsoftheirconstructivework,thatmorethanmereinstinctisinvolvedinit:thecommunityhasindeedarighttoexpectasmuchasthis.(Poincaré1952:xii)
Theresearchprocess:aneight-stepmodel
Researchmethodologyistaughtasasupportingsubjectinseveralwaysinmanyacademicdisciplinesatvariouslevelsbypeoplecommittedtoavarietyofresearchparadigms.Thoughparadigmsvaryintheircontentsandsubstance,theirbroadapproachtoenquiry,intheauthor’sopinion,issimilar.SuchideashavealsobeenexpressedbyFestingerandKatz,whointheforewordoftheirbookResearchMethodsinBehavioralSciencessaythat,‘Althoughthebasiclogicofscientificmethodologyisthesameinallfields,itsspecifictechniquesandapproacheswillvary,dependinguponthesubjectmatter’(1966:vi).Therefore,themodeldevelopedhereisgenericinnatureandcanbeappliedtoanumberofdisciplinesinthesocialsciences.Itisbaseduponapracticalandstep-by-stepapproachtoaresearchenquiryandeachstepprovidesasmorgasbordofmethods,modelsandprocedures.Supposeyouwanttogooutforadrive.Beforeyoustart,youmustdecidewhereyouwanttogoand
thenwhichroutetotake.Ifyouknowtheroute,youdonotneedtoconsultastreetdirectory,but,ifyoudonotknowtheroute,thenyouneedtouseone.Yourproblemiscompoundedifthereismorethanoneroute. You need to decidewhich one to take. The research process is very similar to undertaking ajourney.Aswithyourdrive,foraresearchjourneytherearealsotwoimportantdecisionstomake.Thefirstistodecidewhatyouwanttofindoutaboutor,inotherwords,whatresearchquestionsyouwanttofindanswersto.Havingdecideduponyourresearchquestionsorresearchproblems,youthenneedtodecidehowtogoaboutfindingtheiranswers.Thepathtofindinganswerstoyourresearchquestionsconstitutesresearchmethodology.Justastherearepostsalongthewayasyoutraveltoyourdestination,sotherearepracticalstepsthroughwhichyoumustpassinyourresearchjourneyinordertofindtheanswers to your research questions (Figure 2.1). The sequence of these steps is not fixed and withexperience you can change it. At each operational step in the research process you are required tochoosefromamultiplicityofmethods,proceduresandmodelsofresearchmethodologywhichwillhelpyou best achieve your research objectives. This is where your knowledge base of researchmethodologyplaysacrucialrole.The aim of this book is to provide you with knowledge that will enable you to select the most
appropriate methods and procedures. The strength of this book lies in anchoring the theoreticalknowledgeofthestepsthatyouneedtogothroughonyourresearchjourney.Ateachoperationalstep,thebook aims to provide, at a beginner’s level, knowledgeofmethods andprocedures usedbybothqualitativeandquantitativeresearchers, thoughthereisaninclinationtowardsthequantitativewayofthinking.Quantitativeandqualitativeresearchmethodologiesdifferbothintheirunderpinningphilosophyand,
tosomeextent,inthemethods,modelsandproceduresused.Thoughtheresearchprocessisbroadlythesame in both, quantitative andqualitative research are differentiated in terms of themethods of datacollection,theproceduresadoptedfordataprocessingandanalysis,andthestyleofcommunicationofthefindings.Forexample,ifyourresearchproblemlendsitselftoaqualitativemodeofenquiry,youaremorelikelytousetheunstructuredintervieworobservationasyourmethodofdatacollection.Whenanalysingdatainqualitativeresearch,yougothroughtheprocessofidentifyingthemesanddescribingwhat you have found out during your interviews or observation rather than subjecting your data tostatistical procedures. Table 2.1 summarises the differences between qualitative and quantitativeresearch.
FIGURE2.1Theresearchjourney–toucheachpostandselectmethodsandproceduresappropriateforyourjourney
Since, at a number of steps of the research process, the choice of methods and procedures isinfluencedbythequantitative/qualitativedistinction,themethodsandproceduresdiscussedinsomeofthechaptersinthisbookaredealtwithunderseparateheadingsforqualitativeandquantitativeresearch;however,theauthorhastriedtokeepthisdistinctiontoaminimumasthemodelisapplicabletoboth.Alsonotethatthisbookisforbeginners,itdoesnotcoverextensivelytheapplicabilityanduseofeachmethod,model and procedure. In addition, the author has elaboratedmore onmethods,models andprocedures associated with quantitative research as compared with those linked with qualitativeresearch. For a deeper understanding of a method or procedure relating to either, you may wish toconsultotherbooksidentifiedinthetextorintheBibliography.
TABLE2.1Differencesbetweenqualitativeandquantitativeresearch
Figure2.2 shows theproposedmodel.The tasks identified inarrows are theoperational stepsyouneedtofollowinordertoconductastudy,quantitativeorqualitative.Topicsidentifiedinrectanglesaretherequiredtheoreticalknowledgeneededtocarryoutthesesteps.Thetasksidentifiedincirclesaretheintermediary steps that you need to complete to go from one step to another. It is important for abeginner to work through these steps in the proposed sequence, though, as already stated, withexperienceyoudonotneedtofollowthesequence.Inthisbookthetheoreticalknowledgerequirediswrittenaroundeachoperationalstepandfollows
the samesequentialprogressionas isneededwhenactuallyundertakinga research investigation.Foreach operational step, the required theoretical knowledge is further organised, in different chapters,aroundtheoperationalsteptowhich, in theauthor’sopinion, it ismost logically related(Figure2.3).Again, for abeginner, it is important to study thisdiagram to relate the theoreticalknowledge to theoperationalsteps.
FIGURE2.2Theresearchprocess
FIGURE2.3Thechaptersinthebookinrelationtotheoperationalsteps
Thefollowingsectionsof thischapterprovideaquickglanceat thewholeprocess toacquaintyouwiththevarioustasksyouneedtoundertaketocarryoutyourstudy,thusgivingyousomeideaofwhattheresearchjourneyinvolves.
PhaseI:decidingwhattoresearch
StepI:formulatingaresearchproblem
Formulatingaresearchproblemisthefirstandmostimportantstepintheresearchprocess.Aresearchproblemidentifiesyourdestination:itshouldtellyou,yourresearchsupervisorandyourreaderswhatyouintendtoresearch.Themorespecificandcleareryouarethebetter,aseverythingthatfollowsintheresearchprocess–studydesign,measurementprocedures,samplingstrategy,frameofanalysisandthestyleofwritingofyourdissertationorreport–isgreatlyinfluencedbythewayinwhichyouformulateyour research problem.Hence, you should examine it thoroughly, carefully and critically. Themainfunction of formulating a research problem is to decidewhat youwant to find outabout.Chapter 4dealsindetailwithvariousaspectsofformulatingaresearchproblem.It isextremely important toevaluate the researchproblem in the lightof the financial resourcesat
yourdisposal,thetimeavailable,andyourownandyourresearchsupervisor’sexpertiseandknowledgein the field of study. It is equally important to identify any gaps in your knowledge of relevantdisciplines, such as statistics required for analysis. Also, ask yourself whether you have sufficientknowledgeaboutcomputersandsoftwareifyouplantousethem.
PhaseII:planningaresearchstudy
StepII:conceptualisingaresearchdesign
An extremely important feature of research is the use of appropriate methods. Research involvessystematic, controlled, valid and rigorous exploration and description of what is not known andestablishment of associations and causation that permit the accurate prediction of outcomes under agivensetofconditions.Italsoinvolvesidentifyinggapsinknowledge,verificationofwhatisalreadyknownandidentificationofpasterrorsandlimitations.Thestrengthofwhatyoufindlargelyrestsonhowitwasfound.Themain function of a research design is to explain how youwill find answers to your research
questions.The researchdesignsetsout thespecificdetailsofyourenquiry.Aresearchdesignshouldinclude the following: the study design per se and the logistical arrangements that you propose toundertake, the measurement procedures, the sampling strategy, the frame of analysis and the time-frame. (You should not be confused between study design and research design. Note that the studydesignisonepartoftheresearchdesign.Itisthedesignofthestudyitself,whereastheresearchdesignalsoincludesotherpartswhichconstitutetheresearchprocess.)Foranyinvestigation, theselectionofanappropriateresearchdesign iscrucial inenablingyou to
arriveatvalidfindings,comparisonsandconclusions.Afaultydesignresultsinmisleadingfindingsandisthereforetantamounttowastinghumanandfinancialresources.Inscientificcircles,thestrengthofanempirical investigation is primarily evaluated in the light of the research design adopted. Whenselectingaresearchdesignitisimportanttoensurethatitisvalid,workableandmanageable.Chapter7providesdetailsabouttheresearchdesignmostcommonlyusedinquantitativeandqualitativeresearch.Thereisanenormousvarietyofstudydesignsandyouneedtobeacquaintedwithsomeofthemost
common ones. Chapter 8 explains some of these designs. Select or develop the design that is mostsuitedtoyourstudy.Youmusthavestrongreasonsforselectingaparticulardesign;youmustbeabletojustifyyourselection;andyoushouldbeawareofitsstrengths,weaknessesandlimitations.Inaddition,youwillneedtoexplainthelogisticaldetailsneededtoimplementthesuggesteddesign.
StepIII:constructinganinstrumentfordatacollection
Anythingthatbecomesameansofcollectinginformationforyourstudyiscalleda‘researchtool’ora‘researchinstrument’,forexampleobservationforms,interviewschedules,questionnairesandinterviewguides.Theconstructionofaresearchinstrumentisthefirst‘practical’stepincarryingoutastudy.Youwill
needtodecidehowyouaregoingtocollectdatafortheproposedstudyandthenconstructaresearchinstrumentfordatacollection.Chapter9details thevariousmethodsofdatacollectionforqualitativeandquantitativestudiesandtheprocessofdevelopingaresearchinstrument.If you are planning to collect data specifically for your study (primary data), you need either to
constructaresearchinstrumentortoselectonethathasalreadybeenconstructed.Chapter10dealswith
methodsforcollectingdatausingattitudinalscales.TheconceptsofvalidityandreliabilityinrelationtoaresearchinstrumentarediscussedinChapter11.Ifyouareusingsecondarydata(informationalreadycollectedforotherpurposes),youwillneedto
identifywhat information isneededand thendevelopa form toextract the requireddata. Inorder todeterminewhatinformationisrequired,youneedtogothroughthesameprocessasforprimarydata,describedabove.Fieldtesting(orpre-testing)aresearchtoolisanintegralpartofinstrumentconstruction.Asarule,
thepre-testofaresearchinstrumentshouldnotbecarriedoutonthesampleofyourstudypopulationbutonasimilarpopulationwhichyouarenotproposing tostudy.This iscovered ingreaterdetail inChapter9.Ifyouareplanningtouseacomputerfordataanalysis,youmaywishtoprovidespaceforcodingthe
dataontheresearchinstrument.ThisisexplainedinChapter15.
StepIV:selectingasample
The accuracy of your findings largely depends upon the way you select your sample. The basicobjective of any sampling design is tominimise, within the limitation of cost, the gap between thevaluesobtainedfromyoursampleandthoseprevalentinthestudypopulation.Theunderlyingpremiseinsamplingisthatarelativelysmallnumberofunits,ifselectedinamanner
that they genuinely represent the study population, can provide –with a sufficiently high degree ofprobability–afairlytruereflectionofthesamplingpopulationthatisbeingstudied.Whenselectingasampleyoushouldattempttoachievetwokeyaimsofsamplingtheavoidanceof
bias in the selection of a sample; and the attainment of maximum precision for a given outlay ofresources.There are three categories of sampling design (Chapter 12): random/probability sampling designs,
non-random/non-probabilitysamplingdesignsand‘mixed’samplingdesign.Thereareseveralsamplingstrategieswithinthefirsttwocategories.Youneedtobeacquaintedwith
thesesamplingdesigns–thestrengthsandweaknessesofeachandthesituationsinwhichtheycanorcannotbeapplied– inorder to select theonemostappropriate foryour study.The typeof samplingstrategyyouusewillinfluenceyourabilitytomakegeneralisationsfromthesamplefindingsaboutthestudypopulation,andthetypeofstatisticaltestsyoucanapplytothedata.
StepV:writingaresearchproposal
Havingdoneallthepreparatorywork,thenextstepistoputeverythingtogetherinawaythatprovidesadequate informationaboutyour researchstudy, foryour researchsupervisorandothers.Thisoverallplan,calledaresearchproposal,tellsareaderaboutyourresearchproblemandhowyouareplanningtoinvestigate.Broadly,aresearchproposal’smainfunctionistodetailtheoperationalplanforobtaininganswerstoyourresearchquestions.Indoingsoitensures–andreassuresthereadersof–thevalidityofthemethodologytoobtainanswersaccuratelyandobjectively.Universitiesandotherinstitutionsmayhavedifferingrequirementsregardingthestyleandcontentof
a research proposal, but the majority of institutions would require most of what is set out here.Requirementsmayalsovarywithinan institution, fromdiscipline todisciplineor fromsupervisor tosupervisor.However,theguidelinessetoutinChapter13provideaframeworkwhichwillbeacceptabletomost.Aresearchproposalmusttellyou,yourresearchsupervisorandareviewerthefollowinginformation
aboutyourstudy:
whatyouareproposingtodo;howyouplantoproceed;whyyouselectedtheproposedstrategy.
Thereforeitshouldcontainthefollowinginformationaboutyourstudy(Chapter13):
astatementoftheobjectivesofthestudy;alistofhypotheses,ifyouaretestingany;thestudydesignyouareproposingtouse;thesettingforyourstudy;theresearchinstrument(s)youareplanningtouse;informationonsamplesizeandsamplingdesign;informationondataprocessingprocedures;anoutlineoftheproposedchaptersforthereport;thestudy’sproblemsandlimitations;andtheproposedtime-frame.
PhaseIII:conductingaresearchstudy
StepVI:collectingdata
Havingformulatedaresearchproblem,developedastudydesign,constructedaresearchinstrumentandselectedasample,youthencollect thedatafromwhichyouwilldrawinferencesandconclusionsforyourstudy.Manymethodscouldbeusedtogathertherequiredinformation.Asapartoftheresearchdesign,you
decidedupontheprocedureyouwantedtoadopttocollectyourdata.Inthisphaseyouactuallycollectthe data. For example, depending upon your plans, you might commence interviews, mail out aquestionnaire,conductnominal/focusgroupdiscussionsormakeobservations.Collectingdatathroughanyoneofthemethodsmayinvolvesomeethicalissues,whicharediscussedinChapter14.
StepVII:processinganddisplayingdata
The way you analyse the information you collected largely depends upon two things: the type ofinformation(descriptive,quantitative,qualitativeorattitudinal);andthewayyouwanttocommunicateyourfindingstoyourreaders.Chapter15 describes differentways of analysing quantitative and qualitative data andChapter 16
detailsvariousmethodsofdisplayinganalyseddata.In addition to the qualitative–quantitative distinction, it is important for data analysis that you
considerwhetherthedataistobeanalysedmanuallyorbyacomputer.Ifyourstudyispurelydescriptive,youcanwriteyourdissertation/reportonthebasisofyourfield
notes,manuallyanalysethecontentsofyournotes(contentanalysis),oruseacomputerprogramsuchasNUD*ISTN6,NVivioorEthnographforthispurpose.
Ifyouwantquantitativeanalysis,itisalsonecessarytodecideuponthetypeofanalysisrequired(i.e.frequency distribution, cross-tabulations or other statistical procedures, such as regression analysis,factoranalysisandanalysisofvariance)andhowitshouldbepresented.Youwillalsoneedtoidentifythevariablestobesubjectedtothesestatisticalprocedures.
StepVIII:writingaresearchreport
There are two broad categories of reports: quantitative and qualitative. As mentioned earlier, thedistinction is more academic than real as in most studies you need to combine quantitative andqualitativeskills.Nevertheless,therearesomesolelyqualitativeandsomesolelyquantitativestudies.Writingthereportisthelastand,formany,themostdifficultstepoftheresearchprocess.Thisreport
informs the world what you have done, what you have discovered and what conclusions you havedrawnfromyourfindings.Ifyouareclearaboutthewholeprocess,youwillalsobeclearaboutthewayyouwanttowriteyourreport.Yourreportshouldbewritteninanacademicstyleandbedividedintodifferentchaptersand/orsectionsbaseduponthemainthemesofyourstudy.Chapter17suggestssomeofthewaysofwritingaresearchreport.
SummaryThischapterhasprovidedanoverviewoftheresearchprocess,whichhasbeenbrokendownintoeightsteps,thedetailsofwhicharecoveredintheremainderofthisbook.Ateachsteptheresearchmodelprovidesasmorgasbordofmethods,models,techniquesandproceduressoyoucanselect theonemostappropriateforyourstudy.It is likeabuffetpartywitheighttables,eachwithdifferentdishesmadefromsimilaringredients.Yougotoalleighttablesandselectthedishthatyoulikethemostfromeachtable.Themaindifferencebetweenthemodelandthisexampleisthatinthemodelyouselectwhatismostappropriateforyourstudyandnotwhatyou like the most. For a beginner it is important to go through all the steps, although perhaps not in the same sequence.Withexperienceyoucantakeanumberofshortcuts.Theeightstepscoverthetotalspectrumofaresearchendeavour,fromproblemformulationthroughtowritingaresearchreport.
Thestepsareoperationalinnature,followingalogicalsequence,anddetailingthevariousmethodsandproceduresinasimplestep-by-stepmanner.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Reflectingonthedifferencesbetweenquantitativeandqualitativeresearch(asoutlinedinTable2.1),determinewhichapproachyouaremoreinclinedtofollow.Towhatextentdoesthisreflectyourownunderpinningphilosophy?UsetheinformationprovidedinTable2.1tomapthemaindifferencesbetweenquantitativeandqualitativeresearchateachstepintheeight-stepmodel.
STEPIFormulatingaResearchProblem
Thisoperationalstepincludesfourchapters:
Chapter3:ReviewingtheliteratureChapter4:FormulatingaresearchproblemChapter5:IdentifyingvariablesChapter6:Constructinghypotheses
CHAPTER3ReviewingtheLiterature
Inthischapteryouwilllearnabout:
ThefunctionsoftheliteraturereviewinresearchHowtocarryoutaliteraturesearchHowtoreviewtheselectedliteratureHowtodeveloptheoreticalandconceptualframeworksHowtowritealiteraturereview
Keywords: catalogue, conceptual framework, contextualise, Internet, knowledgebase, literature review, search engines, summary of literature, thematic writing,theoreticalframework.
Theplaceoftheliteraturereviewinresearch
Oneoftheessentialpreliminarytaskswhenyouundertakearesearchstudyistogothroughtheexistingliterature inorder toacquaintyourselfwith theavailablebodyofknowledge inyourareaof interest.Reviewingtheliteraturecanbetimeconsuming,dauntingandfrustrating,butitisalsorewarding.Theliteraturereviewisanintegralpartoftheresearchprocessandmakesavaluablecontributiontoalmostevery operational step. It has value even before the first step; that is,when you aremerely thinkingaboutaresearchquestionthatyoumaywant tofindanswersto throughyourresearchjourney.In theinitialstagesofresearchithelpsyoutoestablishthetheoreticalrootsofyourstudy,clarifyyourideasanddevelopyour researchmethodology.Later in theprocess, the literature reviewserves toenhanceandconsolidateyourownknowledgebaseandhelpsyou to integrateyour findingswith theexistingbodyofknowledge.Sinceanimportantresponsibilityinresearchistocompareyourfindingswiththoseofothers,itisherethattheliteraturereviewplaysanextremelyimportantrole.Duringthewrite-upofyourreportithelpsyoutointegrateyourfindingswithexistingknowledge–thatis,toeithersupportorcontradict earlier research. The higher the academic level of your research, the more important athoroughintegrationofyourfindingswithexistingliteraturebecomes.Insummary,aliteraturereviewhasthefollowingfunctions:
Itprovidesatheoreticalbackgroundtoyourstudy.Ithelpsyouestablishthelinksbetweenwhatyouareproposingtoexamineandwhathasalreadybeenstudied.Itenablesyoutoshowhowyourfindingshavecontributedtotheexistingbodyofknowledgeinyourprofession.Ithelpsyoutointegrateyourresearchfindingsintotheexistingbodyofknowledge.
Inrelationtoyourownstudy,theliteraturereviewcanhelpinfourways.Itcan:
1. bringclarityandfocustoyourresearchproblem;2. improveyourresearchmethodology;3. broadenyourknowledgebaseinyourresearcharea;and4. contextualiseyourfindings.
Bringingclarityandfocustoyourresearchproblem
Theliteraturereviewinvolvesaparadox.Ontheonehand,youcannoteffectivelyundertakealiteraturesearch without some idea of the problem you wish to investigate. On the other hand, the literaturereviewcanplayanextremelyimportantroleinshapingyourresearchproblembecausetheprocessofreviewing the literature helps you to understand the subject area better and thus helps you toconceptualiseyourresearchproblemclearlyandpreciselyandmakesitmorerelevantandpertinenttoyourfieldofenquiry.Whenreviewingtheliteratureyoulearnwhataspectsofyoursubjectareahavebeenexaminedbyothers,whattheyhavefoundoutabouttheseaspects,whatgapstheyhaveidentifiedandwhatsuggestionstheyhavemadeforfurtherresearch.Allthesewillhelpyougainagreaterinsightintoyourownresearchquestionsandprovideyouwithclarityandfocuswhicharecentraltoarelevantandvalidstudy.Inaddition, itwillhelpyoutofocusyourstudyonareaswheretherearegapsintheexistingbodyofknowledge,therebyenhancingitsrelevance.
Improvingyourresearchmethodology
Goingthroughtheliteratureacquaintsyouwiththemethodologiesthathavebeenusedbyotherstofindanswers toresearchquestionssimilar to theoneyouare investigating.Aliteraturereviewtellsyou ifothershaveusedproceduresandmethodssimilartotheonesthatyouareproposing,whichproceduresandmethodshaveworkedwellforthemandwhatproblemstheyhavefacedwiththem.Bybecomingawareofanyproblemsandpitfalls,youwillbebetterpositionedtoselectamethodologythatiscapableof providing valid answers to your research question. This will increase your confidence in themethodologyyouplantouseandwillequipyoutodefenditsuse.
Broadeningyourknowledgebaseinyourresearcharea
Themost important functionof the literature review is to ensureyou readwidelyaround the subjectarea in which you intend to conduct your research study. It is important that you knowwhat otherresearchershavefoundinregardtothesameorsimilarquestions,whattheorieshavebeenputforwardandwhatgapsexist in the relevantbodyofknowledge.Whenyouundertakea researchproject forahigher degree (e.g. anMA or a PhD) you are expected to be an expert in your area of research.A
thorough literature review helps you to fulfil this expectation.Another important reason for doing aliteraturereviewis that ithelpsyoutounderstandhowthefindingsofyourstudyfit intotheexistingbodyofknowledge(Martin1985:30).
Enablingyoutocontextualiseyourfindings
Obtaininganswerstoyourresearchquestionsiscomparativelyeasy:thedifficultpartisexamininghowyour findings fit into the existing body of knowledge. How do answers to your research questionscomparewithwhatothershavefound?Whatcontributionhaveyoubeenabletomaketotheexistingbody of knowledge?How are your findings different from those of others?Undertaking a literaturereviewwillenableyoutocompareyourfindingswiththoseofothersandanswerthesequestions.Itisimportanttoplaceyourfindingsinthecontextofwhatisalreadyknowninyourfieldofenquiry.
Howtoreviewtheliterature
Ifyoudonothaveaspecificresearchproblem,youshouldreviewtheliteratureinyourbroadareaofinterestwiththeaimofgraduallynarrowingitdowntowhatyouwanttofindoutabout.Afterthattheliteraturereviewshouldbefocusedaroundyourresearchproblem.Thereisadangerinreviewingtheliteraturewithouthavinga reasonably specific ideaofwhatyouwant to study. It canconditionyourthinkingaboutyourstudyandthemethodologyyoumightuse,resultinginalessinnovativechoiceofresearchproblemandmethodology thanotherwisewouldhavebeen the case.Hence, you should trybroadlytoconceptualiseyourresearchproblembeforeundertakingyourmajorliteraturereview.Therearefourstepsinvolvedinconductingaliteraturereview:
1. Searchingfortheexistingliteratureinyourareaofstudy.2. Reviewingtheselectedliterature.3. Developingatheoreticalframework.4. Developingaconceptualframework.
Theskillsrequiredforthesetasksaredifferent.Developingtheoreticalandconceptualframeworksismoredifficultthantheothertasks.
Searchingfortheexistingliterature
Tosearcheffectively for the literature inyour fieldofenquiry, it is imperative thatyouhaveat leastsome idea of the broad subject area and of the problem you wish to investigate, in order to setparameters foryoursearch.Next,compileabibliographyfor thisbroadarea.Thereare threesourcesthatyoucanusetoprepareabibliography:
(a)books;(b)journals;(c)theInternet.
Books
Though books are a central part of any bibliography, they have their disadvantages as well asadvantages.Themainadvantageisthatthematerialpublishedinbooksisusuallyimportantandofgoodquality, and the findings are ‘integratedwith other research to form a coherent body of knowledge’(Martin1985:33).Themaindisadvantageisthatthematerialisnotcompletelyuptodate,asitcantakeafewyearsbetweenthecompletionofaworkanditspublicationintheformofabook.Thebestwaytosearchforabookistolookatyourlibrarycatalogues.Whenlibrarianscataloguea
book they also assign to it subject headings that are usually based on Library of Congress SubjectHeadings.Ifyouarenotsure,askyourlibrariantohelpyoufindthebestsubjectheadingforyourarea.Thiscansaveyoualotoftime.PublicationssuchasBookReviewIndexcanhelpyoutolocatebooksofinterest.Usethesubjectcatalogueorkeywordsoptiontosearchforbooksinyourareaofinterest.Narrowthe
subject area searched by selecting the appropriate keywords. Look through these titles carefully andidentifythebooksyouthinkarelikelytobeofinteresttoyou.Ifyouthinkthetitlesseemappropriatetoyourtopic,printthemout(ifthisfacilityisavailable),asthiswillsaveyoutime,ornotethemdownona piece of paper. Be aware that sometimes a title does not provide enough information to help youdecideifabookisgoingtobeofusesoyoumayhavetoexamineitscontentstoo.When you have selected 10–15 books that you think are appropriate for your topic, examine the
bibliography of each one. It will save time if you photocopy their bibliographies. Go through thesebibliographiescarefullytoidentifythebookscommontoseveralofthem.Ifabookhasbeenreferencedbyanumberofauthors,youshouldincludeitinyourreadinglist.Prepareafinallistofbooksthatyouconsideressentialreading.Having prepared your reading list, locate these books in your library or borrow them from other
sources.Examinetheircontentstodouble-checkthattheyreallyarerelevanttoyourtopic.Ifyoufindthatabookisnotrelevanttoyourresearch,deleteitfromyourreadinglist.Ifyoufindthatsomethinginabook’scontentsisrelevanttoyourtopic,makeanannotatedbibliography.Anannotatedbibliographycontainsabriefabstractoftheaspectscoveredinabookandyourownnotesofitsrelevance.Becarefultokeep trackofyour references.Todo thisyoucanprepareyourowncard indexoruseacomputerprogramsuchasEndnotesorPro-Cite.
Journals
Youneedtogothroughthejournalsrelatingtoyourresearchinasimilarmanner.Journalsprovideyouwiththemostup-to-dateinformation,eventhoughthereisoftenagapoftwotothreeyearsbetweenthecompletionofaresearchprojectanditspublicationinajournal.Youshouldselectasmanyjournalsasyoupossibly can, though the number of journals available depends upon the field of study– certainfieldshavemorejournalsthanothers.Aswithbooks,youneedtopreparealistofthejournalsyouwanttoexamineforidentifyingtheliteraturerelevanttoyourstudy.Thiscanbedoneinanumberofways.Youcan:
locatethehardcopiesofthejournalsthatareappropriatetoyourstudy;lookatcitationorabstractindicestoidentifyand/orreadtheabstractsofsucharticles;searchelectronicdatabases.
Ifyouhavebeenabletoidentifyanyusefuljournalsandarticles,preparealistofthoseyouwanttoexamine,byjournal.Selectoneofthesejournalsand,startingwiththelatestissue,examineitscontentspagetoseeifthereisanarticleofrelevancetoyourresearchtopic.Ifyoufeelthataparticulararticleisofinteresttoyou,readitsabstract.Ifyouthinkyouarelikelytouseit,dependinguponyourfinancialresources,eitherphotocopyit,orprepareasummaryandrecorditsreferenceforlateruse.Thereareseveralsourcesdesigned tomakeyoursearch for journalseasierand thesecansaveyou
enormoustime.Theyare:
indicesofjournals(e.g.HumanitiesIndex);abstractsofarticles(e.g.ERIC);citationindices(e.g.SocialSciencesCitationIndex).
Eachof these indexing,abstractingandcitationservices isavailable inprint,oraccessible throughtheInternet.Inmostlibraries, informationonbooks, journalsandabstractsisstoredoncomputers.Ineachcase
the information is classified by subject, author and title. You may also have the keywords option(author/keyword;title/keyword;subject/keyword;expert/keyword;orjustkeywords).Whatsystemyouusedependsuponwhatisavailableinyourlibraryandwhatyouarefamiliarwith.There are specially prepared electronic databases in a number of disciplines. These can also be
helpfulinpreparingabibliography.Forexample,mostlibrariescarrytheelectronicdatabasesshowninTable3.1.Selectthedatabasemostappropriatetoyourareaofstudytoseeifthereareanyusefulreferences.Of
course,anycomputerdatabasesearchisrestrictedtothosejournalsandarticlesthatarealreadyonthedatabase.Youshouldalsotalktoyourresearchsupervisorandotheravailableexpertstofindoutaboutanyadditionalrelevantliteraturetoincludeinyourreadinglist.
TABLE3.1Somecommonlyusedelectronicdatabasesinpublichealth,sociology,educationandbusinessstudies
TheInternet
Inalmosteveryacademicdisciplineandprofessional field, theInternethasbecomean important toolfor finding published literature. Through an Internet search you can identify published material inbooks,journalsandothersourceswithimmenseeaseandspeed.AnInternetsearchiscarriedoutthroughsearchengines,ofwhichtherearemany,thoughthemost
commonlyusedareGoogleandYahoo.SearchingthroughtheInternetisverysimilartothesearchforbooksandarticlesinalibraryusinganelectroniccatalogue,asitisbasedontheuseofkeywords.AnInternet search basically identifies all material in the database of a search engine that contains thekeywordsyouspecify,either individuallyor incombination.It is important thatyouchoosewordsorcombinationsofwordsthatotherpeoplearelikelytouse.AccordingtoGilbert(2008:73),‘MostsearchfacilitiesuseBooleanlogic,whichallowsthreetypes
ofbasicsearch“AND”,“OR”and“NOT”.’WithpracticeyouwillbecomemoreefficientandeffectiveinusingkeywordsincombinationwithAND,ORandNOT,andsolearntonarrowyoursearchtohelpyouidentifythemostrelevantreferences.
Reviewingtheselectedliterature
Nowthatyouhaveidentifiedseveralbooksandarticlesasuseful,thenextstepistostartreadingthemcritically to pull together themes and issues that are of relevance to your study. Unless you have atheoreticalframeworkof themesinmindtostartwith,useseparatesheetsofpaperforeachthemeorissue you identify as you go through selected books and articles. The following example details theprocess.
Theauthorrecentlyexamined,aspartofanevaluationstudy,theextentofpracticeoftheconceptof ‘community responsiveness’ in the deliveryof health services inWesternAustralia byhealthservice providers. Before evaluating the extent of its use, pertinent literature relating to‘community responsiveness in health’ was identified and reviewed. Through this review, manythemesemerged,whichbecame thebasisofdeveloping the theoretical framework for the study.Outofallofthis,thefollowingthemeswereselectedtoconstructthetheoreticalframeworkfortheevaluationstudy:
Communityresponsiveness:whatdoesitmean?Philosophiesunderpinningcommunityresponsiveness.HistoricaldevelopmentoftheconceptinAustralia.Theextentofuseinhealthplanning?Strategiesdevelopedtoachievecommunityresponsiveness.Indicatorsofsuccessorfailure.Seekingcommunityparticipation.Difficultiesinimplementingcommunityresponsiveness.Attitudeofstakeholderstowardstheconceptofcommunityresponsiveness.
Onceyoudevelop a rough framework, slot the findings from thematerial so far reviewed into these
themes,usingaseparatesheetofpaperforeachthemeoftheframeworksofardeveloped.Asyoureadfurther, goon slotting the informationwhere it logicallybelongsunder the themes so far developed.Keep in mind that you may need to add more themes as you go along. While going through theliteratureyoushouldcarefullyandcriticallyexamineitwithrespecttothefollowingaspects:
Notewhethertheknowledgerelevanttoyourtheoreticalframeworkhasbeenconfirmedbeyonddoubt.Notethetheoriesputforward,thecriticismsoftheseandtheirbasis,themethodologiesadopted(studydesign,samplesizeanditscharacteristics,measurementprocedures,etc.)andthecriticismsofthem.Examinetowhatextentthefindingscanbegeneralisedtoothersituations.Noticewheretherearesignificantdifferencesofopinionamongresearchersandgiveyouropinionaboutthevalidityofthesedifferences.Ascertaintheareasinwhichlittleornothingisknown–thegapsthatexistinthebodyofknowledge.
Developingatheoreticalframework
Examiningtheliteraturecanbeanever-endingtask,butasyouhavelimitedtimeitisimportanttosetparametersbyreviewingtheliteratureinrelationtosomemainthemespertinenttoyourresearchtopic.Asyoustartreadingtheliterature,youwillsoondiscoverthattheproblemyouwishtoinvestigatehasitsrootsinanumberoftheoriesthathavebeendevelopedfromdifferentperspectives.Theinformationobtainedfromdifferentbooksandjournalsnowneedstobesortedunderthemainthemesandtheories,highlighting agreements and disagreements among the authors and identifying the unansweredquestionsorgaps.Youwill also realise that the literaturedealswithanumberofaspects thathaveadirect or indirect bearing on your research topic. Use these aspects as a basis for developing yourtheoretical framework. Your review of the literature should sort out the information, as mentionedearlier,withinthisframework.Unlessyoureviewtheliteratureinrelationtothisframework,youwillnotbeabletodevelopafocusinyourliteraturesearch:thatis,yourtheoreticalframeworkprovidesyouwithaguideasyouread.Thisbringsustotheparadoxmentionedpreviously:untilyougothroughtheliterature you cannot develop a theoretical framework, and until you have developed a theoreticalframeworkyoucannoteffectivelyreviewtheliterature.Thesolutionistoreadsomeoftheliteratureandthenattempttodevelopaframework,evenalooseone,withinwhichyoucanorganisetherestoftheliteratureyouread.Asyoureadmoreaboutthearea,youarelikelytochangetheframework.However,withoutit,youwillgetboggeddowninagreatdealofunnecessaryreadingandnote-takingthatmaynotberelevanttoyourstudy.Literaturepertinenttoyourstudymaydealwithtwotypesofinformation:
1. universal;2. morespecific(i.e.localtrendsoraspecificprogramme).
Inwritingaboutsuchinformationyoushouldstartwiththegeneralinformation,graduallynarrowingitdowntothespecific.LookattheexampleinFigure3.1aand3.1b
FIGURE3.1aDevelopingatheoreticalframework–therelationshipbetweenmortalityandfertility
FIGURE3.1bTheoreticalframeworkforthestudy‘communityresponsivenessinhealth’
Developingaconceptualframework
The conceptual framework is the basis of your research problem. It stems from the theoreticalframeworkandusually focuseson the section(s)whichbecome thebasisofyour study.Whereas thetheoreticalframeworkconsistsofthetheoriesorissuesinwhichyourstudyisembedded,theconceptualframeworkdescribes the aspectsyou selected from the theoretical framework tobecome thebasisofyourenquiry.Forinstance,intheexamplecitedinFigure3.1a, thetheoreticalframeworkincludesallthe theories that have been put forward to explain the relationship between fertility and mortality.However,outofthese,youmaybeplanningtotestonlyone,saythefearofnon-survival.Similarly,inFigure3.1b,theconceptualframeworkisfocusedonindicatorstomeasurethesuccessorfailureofthestrategies to enhance community responsiveness. Hence the conceptual framework grows out of thetheoreticalframeworkandrelatestothespecificresearchproblem.
Writingabouttheliteraturereviewed
Now,allthatremainstobedoneistowriteabouttheliteratureyouhavereviewed.Asmentionedinthebeginning of this chapter, two of the broad functions of a literature review are (1) to provide atheoreticalbackgroundtoyourstudyand(2)toenableyoutocontextualiseyourfindingsinrelationtotheexistingbodyofknowledgeinadditiontorefiningyourmethodology.Thecontentofyourliteraturereviewshould reflect these twopurposes. Inorder to fulfil the firstpurpose,youshould identifyanddescribe various theories relevant to your field; and specify gaps in existing knowledge in the area,recent advances in the area of study, current trends and so on. In order to comply with the secondfunctionyoushouldintegratetheresultsfromyourstudywithspecificandrelevantfindingsfromtheexistingliteraturebycomparingthe twoforconfirmationorcontradiction.Note thatat thisstageyoucanonlyaccomplish the first functionof the literature review, toprovidea theoreticalbackground toyourstudy.Forthesecondfunction,thecontextualisationofthefindings,youhavetowaittillyouareattheresearchreportwritingstage.While reading the literature for theoretical background of your study, youwill realise that certain
themeshaveemerged.Listthemainones,convertingthemintosubheadings.Somepeoplewriteuptheentire literature review in one section, entitled ‘Review of the literature’, ‘Summary of literature’ or‘The literature review’, without subheadings, but the author strongly suggests that you write yourliteraturereviewundersubheadingsbaseduponthemainthemesthatyouhavediscoveredandwhichformthebasisofyourtheoreticalframework.Thesesubheadingsshouldbeprecise,descriptiveofthetheme in question and follow a logical progression. Now, under each subheading, record the mainfindings with respect to the theme in question (thematic writing), highlighting the reasons for andagainst anargument if theyexist, and identifyinggaps and issues.Figure3.2 shows the subheadingsused to describe the themes in a literature review conducted by the author for a study entitled‘IntercountryadoptioninWesternAustralia’.
FIGURE3.2Sampleofoutlineofaliteraturereview
The second broad function of the literature review – contextualising the findings of your study –requiresyoutocompareverysystematicallyyourfindingswiththosemadebyothers.Quotefromthesestudies to showhowyour findings contradict, confirmor add to them. It placesyour findings in thecontext of what others have found out providing complete reference in an acceptable format. Thisfunctionisundertaken,asmentionedearlier,whenwritingaboutyourfindings,thatisafteranalysisofyourdata.
SummaryReviewing the literature is a continuousprocess. It beginsbefore a researchproblem is finalised and continuesuntil the report isfinished.Thereisaparadoxintheliteraturereview:youcannotundertakeaneffectiveliteraturereviewunlessyouhaveformulatedaresearchproblem,yetyourliteraturesearchplaysanextremelyimportantroleinhelpingyoutoformulateyourresearchproblem.Theliterature review brings clarity and focus to your research problem, improves your research methodology and broadens yourknowledgebase.Reviewingtheliteratureinvolvesanumberofsteps:searchingforexistingliteratureinyourareaofstudy;reviewingtheselected
literature; using it to develop a theoretical framework fromwhich your study emerges and also using it to develop a conceptualframeworkwhichwillbecomethebasisofyourinvestigation.Themainsourcesforidentifyingliteraturearebooks,journalsandtheInternet.Thereareseveralsourceswhichcanprovideinformationaboutlocatingrelevantjournals.Theliteraturereviewservestwoimportantfunction:(1)itprovidestheoreticalbackgroundtoyourstudy,and(2)ithelpsyouto
contextualiseyourfindingsbycomparingthemwithwhatothershavefoundoutinrelationtotheareaofenquiry.Atthisstageoftheresearch process, only the first function can be fulfilled.You can only take steps to achieve the second functionwhen you haveanalysedyourdataandareintheprocessofwritingaboutyourfindings.Yourwritingabouttheliteraturereviewedshouldbethematicinnature,thatisbasedonmainthemes;thesequenceofthesethemes
inthewrite-upshouldfollowalogicalprogression;variousargumentsshouldbesubstantiatedwithspecificquotationsandcitationsfromtheliteratureandshouldadheretoanacceptableacademicreferencingstyle.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Undertakeakeywordsearchforathemeorissuethatinterestsyouusing(a)anInternetsearchengine,suchasGoogleScholar,and(b)alibrarysearchfacility.Comparetheresults.Choosetwoorthreeresearchreportsfromyoursearchandscanthroughthesummariesnotingthetheoriesputforward,themethodologiesadoptedandanyrecommendationsforfurtherstudy.Dothesereportspointtoaconsensusordifferencesofopinioninthefield?Developatheoreticalframeworkforthethemeorissueyouselected.
CHAPTER4FormulatingaResearchProblem
Inthischapteryouwilllearnabout:
TheimportanceofformulatingaresearchproblemSourcesofresearchproblemsConsiderationsinselectingaresearchproblemSpecificissuestoconsiderwhenformulatingaresearchprobleminqualitativeresearchStepsinformulatingaresearchproblemHowtoformulateresearchobjectivesTheimportanceofestablishingoperationaldefinitions
Keywords: concepts, dissect, operational definition, qualitative research,quantitative research, research objectives, research problem, study area, studypopulation,subjectarea,validity,variable,workingdefinition.
Thecentralaimofthischapteristodetailtheprocessofformulatingaresearchproblem,eventhoughthespecificprocessthatyouarelikelytoadoptdependsupon:
yourexpertiseinresearchmethodology;yourknowledgeofthesubjectarea;yourunderstandingoftheissuestobeexamined;theextenttowhichthefocusofyourstudyispredetermined.
Ifyouarenotveryfamiliarwiththeresearchprocessand/ordonothaveaveryspecificideaaboutwhat is to be researched, you need to follow every step detailed in this chapter. However, moreexperienced researchers can take a number of shortcuts.The process outlined here assumes that youhaveneither the requiredknowledgeof theprocessof formulating a researchproblemnor a specificideaaboutwhatistoberesearched.Ifyouhaveaspecificideaforthebasisofyourenquiry,youdonotneedtogothroughthischapter.However,youshouldmakesurethatyourideaisresearchableasnotallproblemslendthemselvestoresearchmethodologies.
Theresearchproblem
Broadlyspeaking,anyquestionthatyouwantansweredandanyassumptionorassertionthatyouwanttochallengeorinvestigatecanbecomearesearchproblemoraresearchtopicforyourstudy.However,itisimportanttorememberthatnotallquestionscanbetransformedintoresearchproblemsandsomemayprove tobeextremelydifficult tostudy.According toPowers,MeenaghanandTwoomey(1985:38), ‘Potential researchquestionsmayoccur to us on a regular basis, but theprocess of formulatingtheminameaningfulwayisnotatallaneasytask.’Asanewcomeritmightseemeasytoformulateaproblembut it requires considerable knowledge of both the subjectarea and researchmethodology.Onceyouexamineaquestionmorecloselyyouwillsoonrealisethecomplexityofformulatinganideaintoaproblemwhichisresearchable.‘First identifyingandthenspecifyingaresearchproblemmightseemlikeresearchtasksthatoughttobeeasyandquicklyaccomplished.However,suchisoftennotthecase’(Yegidis&Weinback1991:35).It is essential for the problem you formulate to be able to withstand scrutiny in terms of the
proceduresrequiredtobeundertaken.Henceyoushouldspendconsiderabletimeinthinkingitthrough.
Theimportanceofformulatingaresearchproblem
Theformulationofaresearchproblemisthefirstandmostimportantstepoftheresearchprocess.Itisliketheidentificationofadestinationbeforeundertakingajourney.Intheabsenceofadestination,itisimpossibletoidentifytheshortest–orindeedany–route.Similarly,intheabsenceofaclearresearchproblem,aclearandeconomicalplanisimpossible.Touseanotheranalogy,aresearchproblemislikethefoundationofabuilding.Thetypeanddesignofthebuildingaredependentuponthefoundation.Ifthefoundationiswelldesignedandstrongyoucanexpectthebuildingtobealso.Theresearchproblemservesas the foundationofa researchstudy: if it iswell formulated,youcanexpectagoodstudy tofollow.AccordingtoKerlinger:
Ifonewantstosolveaproblem,onemustgenerallyknowwhattheproblemis.Itcanbesaidthatalargepartoftheproblemliesinknowingwhatoneistryingtodo.(1986:17)
Youmusthaveaclearideawithregardtowhatitisthatyouwanttofindoutaboutandnotwhatyouthinkyoumustfind.Aresearchproblemmaytakeanumberofforms,fromtheverysimpletotheverycomplex.Theway
youformulateaproblemdeterminesalmosteverystepthatfollows:thetypeofstudydesignthatcanbeused;thetypeofsamplingstrategythatcanbeemployed;theresearchinstrumentthatcanbeusedordeveloped; and the type of analysis that can be undertaken. Suppose your broad area of interest isdepression. Further suppose you want to conduct a research study regarding services available topatientswithdepressionlivinginacommunity.Ifyourfocusistofindoutthetypesofserviceavailabletopatientswithdepression, the studywill dominantlybedescriptive andqualitative innature.Thesetypesofstudiesfallinthecategoryofqualitativeresearchandarecarriedoutusingqualitativeresearchmethodologies.Ontheotherhand,ifyouwanttofindouttheextentofuseoftheseservices,thatisthenumber of people using them, it will dominantly use quantitative methodologies even though it isdescriptiveinnaturedescribingthenumberofpeopleusingaservice.Ifyourfocusistodeterminetheextent of use in relation to the personal attributes of the patients, the study will be classified ascorrelational(andquantitative).Themethodologyusedwillbedifferentthantheoneusedinthecaseofadescriptivestudy.Similarly,ifyouraimistofindouttheeffectivenessoftheseservices,thestudywill
againbeclassifiedascorrelationalandthestudydesignused,methodsofcollectingdataanditsanalysiswillbeapartofthequantitativemethodology.Hence,itisimportantforyoutounderstandthatthewayyou formulatea researchproblemdeterminesall thesubsequent steps thatyouhave to followduringyourresearchjourney.The formulation of a problem is like the ‘input’ to a study, and the ‘output’ – the quality of the
contentsoftheresearchreportandthevalidityoftheassociationsorcausationestablished–isentirelydependent upon it. Hence the famous saying about computers, ‘garbage in, garbage out’, is equallyapplicabletoaresearchproblem.Initially,youmaybecomemoreconfusedbut this isnormalandasignofprogression.Remember:
confusion is often but a first step towards clarity. Take time over formulating your problem, for thecleareryouareaboutyourresearchproblem/question,theeasieritwillbeforyoulateron.Remember,thisisthemostcrucialstep.
Sourcesofresearchproblems
This section isofparticular relevance ifyouhavenotyet selecteda research topic anddonotknowwheretostart.Ifyouhavealreadyselectedyourtopicorquestion,gotothenextsection.MostresearchinthehumanitiesrevolvesaroundfourPs:
people;problems;programmes;phenomena.
In fact, a closer look at any academic or occupational fieldwill show thatmost research revolvesaroundthesefourPs.Theemphasisonaparticular‘P’mayvaryfromstudytostudybutgenerally,inpractice,mostresearchstudiesarebaseduponatleastacombinationoftwoPs.Youmayselectagroupofindividuals(agroupofindividuals–oracommunityassuch–‘people’),toexaminetheexistenceofcertainissuesorproblemsrelatingtotheirlives,toascertaintheirattitudetowardsanissue(‘problem’),to establish the existence of a regularity (‘phenomenon’) or to evaluate the effectiveness of anintervention(‘programme’).Yourfocusmaybethestudyofanissue,anassociationoraphenomenonperse;forexample,therelationshipbetweenunemploymentandstreetcrime,smokingandcancer,orfertility andmortality,which is done on the basis of information collected from individuals, groups,communitiesororganisations.Theemphasisinthesestudiesisonexploring,discoveringorestablishingassociationsorcausation.Similarly,youcanstudydifferentaspectsofaprogramme:itseffectiveness,itsstructure,theneedforit,consumers’satisfactionwithit,andsoon.Inordertoascertaintheseyoucollectinformationfrompeople.Everyresearchstudyhas twoaspects: thepeopleprovideyouwith the‘studypopulation’,whereas
the other three Ps furnish the ‘subject areas’. Your study population – individuals, groups andcommunities–is thepeople fromwhomthe information iscollected.Yoursubjectarea isaproblem,programmeorphenomenonaboutwhichtheinformationiscollected.ThisisoutlinedfurtherinTable4.1,whichshowstheaspectsofaresearchproblem.
TABLE4.1Aspectsofaresearchproblem
You can study a problem, a programme or a phenomenon in any academic field or from anyprofessionalperspective.Forexample,youcanmeasuretheeffectivenessofaprogrammeinthefieldofhealth, education, social work, industrial management, public health, nursing, health promotion orwelfare,oryoucanlookataproblemfromahealth,businessorwelfareperspective.Similarlyyoucangaugeconsumers’opinionsaboutanyaspectofaprogrammeintheabovefields.ExamineyourownacademicdisciplineorprofessionalfieldinthecontextofthefourPsinorderto
identifyanythingthatlooksinteresting.Forexample,ifyouareastudentinthehealthfieldthereareanenormousnumberof issues, situationsandassociationswithineach subfieldofhealth thatyoucouldexamine.Issuesrelatingtothespreadofadisease,drugrehabilitation,animmunisationprogramme,theeffectiveness of a treatment, the extent of consumers’ satisfaction or issues concerning a particularhealthprogrammecanallprovideyouwitharangeofresearchproblems.Similarly,ineducationthereareseveral issues:students’satisfactionwitha teacher,attributesofagoodteacher, the impactof thehome environment on the educational achievement of students, and the supervisory needs ofpostgraduate students inhighereducation.Anyotheracademicoroccupational fieldcan similarlybedissectedintosubfieldsandexaminedforapotentialresearchproblem.Mostfieldslendthemselvestotheabovecategorisationeven thoughspecificproblemsandprogrammesvarymarkedlyfromfield tofield.The concept of 4Ps is applicable to both quantitative and qualitative research though the main
differenceat this stage is theextentof their specificity,dissection,precisionand focus. Inqualitativeresearchtheseattributesaredeliberatelykeptveryloosesothatyoucanexploremoreasyougoalong,incaseyoufindsomethingofrelevance.Youdonotbindyourselfwithconstraintsthatwouldputlimitsonyourabilitytoexplore.Thereisaseparatesectionon‘Formulatingaresearchprobleminqualitativeresearch’laterinthechapter,whichprovidesfurtherguidanceontheprocess.
Considerationsinselectingaresearchproblem
Whenselectingaresearchproblem/topic thereareanumberofconsiderations tokeep inmindwhichwill help to ensure that your study will be manageable and that you remain motivated. Theseconsiderationsare:
Interest–Interestshouldbethemostimportantconsiderationinselectingaresearchproblem.Aresearchendeavourisusuallytimeconsuming,andinvolveshardworkandpossiblyunforeseenproblems.Ifyouselectatopicwhichdoesnotgreatlyinterestyou,itcouldbecomeextremelydifficulttosustaintherequiredmotivationandputinenoughtimeandenergytocompleteit.Magnitude–Youshouldhavesufficientknowledgeabouttheresearchprocesstobeabletovisualisetheworkinvolvedincompletingtheproposedstudy.Narrowthetopicdowntosomethingmanageable,specificandclear.Itisextremelyimportanttoselectatopicthatyoucanmanage
withinthetimeandwiththeresourcesatyourdisposal.Evenifyouareundertakingadescriptivestudy,youneedtoconsideritsmagnitudecarefully.Measurementofconcepts–Ifyouareusingaconceptinyourstudy(inquantitativestudies),makesureyouareclearaboutitsindicatorsandtheirmeasurement.Forexample,ifyouplantomeasuretheeffectivenessofahealthpromotionprogramme,youmustbeclearastowhatdetermineseffectivenessandhowitwillbemeasured.Donotuseconceptsinyourresearchproblemthatyouarenotsurehowtomeasure.Thisdoesnotmeanyoucannotdevelopameasurementprocedureasthestudyprogresses.Whilemostofthedevelopmentalworkwillbedoneduringyourstudy,itisimperativethatyouarereasonablyclearaboutthemeasurementoftheseconceptsatthisstage.Levelofexpertise–Makesureyouhaveanadequatelevelofexpertiseforthetaskyouareproposing.Allowforthefactthatyouwilllearnduringthestudyandmayreceivehelpfromyourresearchsupervisorandothers,butrememberthatyouneedtodomostoftheworkyourself.Relevance–Selectatopicthatisofrelevancetoyouasaprofessional.Ensurethatyourstudyaddstotheexistingbodyofknowledge,bridgescurrentgapsorisusefulinpolicyformulation.Thiswillhelpyoutosustaininterestinthestudy.Availabilityofdata–Ifyourtopicentailscollectionofinformationfromsecondarysources(officerecords,clientrecords,censusorotheralready-publishedreports,etc.)makesurethatthisdataisavailableandintheformatyouwantbeforefinalisingyourtopic.Ethicalissues–Anotherimportantconsiderationinformulatingaresearchproblemistheethicalissuesinvolved.Inthecourseofconductingaresearchstudy,thestudypopulationmaybeadverselyaffectedbysomeofthequestions(directlyorindirectly);deprivedofanintervention;expectedtosharesensitiveandprivateinformation;orexpectedtobesimplyexperimental‘guineapigs’.Howethicalissuescanaffectthestudypopulationandhowethicalproblemscanbeovercomeshouldbethoroughlyexaminedattheproblem-formulationstage.
Stepsinformulatingaresearchproblem
Theformulationofaresearchproblemisthemostcrucialpartoftheresearchjourneyasthequalityandrelevance of your research project entirely depends upon it. As mentioned earlier, every step thatconstitutesthehowpartoftheresearchjourney(Figure2.1)dependsuponthewayyouformulatedyourresearchproblem.Despite theimportanceof thisstep, thereisverylittleavailablebywayofspecificguidance inotherbooks.This task is largely lefteither to the teachersof researchmethodologyor tostudents to learn for themselves.Oneof the strengthsof this book is that it offers abeginner averyspecificsetofstep-by-stepguidelinesinoneplacedespitethefearofbeinglabelledasprescriptive.Theprocessofformulatingaresearchproblemconsistsofanumberofsteps.Workingthroughthese
stepspresupposesareasonablelevelofknowledgeinthebroadsubjectareawithinwhichthestudyistobe undertaken and the research methodology itself. A brief review of the relevant literature helpsenormously in broadening this knowledge base.Without such knowledge it is difficult to ‘dissect’ asubjectareaclearlyandadequately.Ifyoudonotknowwhatspecificresearchtopic,idea,questionsorissueyouwanttoresearch(which
isnotuncommonamongstudents),firstgothroughthefollowingsteps:
FIGURE4.1Dissectingthesubjectareaofdomesticviolenceintosubareas
Step1
Identifyabroadfieldorsubjectareaofinteresttoyou.Askyourself,‘Whatisitthatreallyinterestsmeasaprofessional?’Intheauthor’sopinion,itisagoodideatothinkaboutthefieldinwhichyouwouldliketoworkaftergraduation.Thiswillhelpyoutofindaninterestingtopic,andonewhichmaybeofusetoyouinthefuture.Forexample,ifyouareasocialworkstudent, inclined towork in the area of youthwelfare, refugees or domestic violence after graduation, youmight take toresearchinoneoftheseareas.Orifyouarestudyingmarketingyoumightbeinterestedinresearchingconsumerbehaviour.Or,asastudentofpublichealth,intendingtoworkwithpatientswhohaveHIV/AIDS,youmightliketoconductresearchonasubjectarearelatingtoHIV/AIDS.Asfarastheresearchjourneygoes,thesearethebroadresearchareas.Itisimperativethatyouidentifyoneofinteresttoyoubeforeundertakingyourresearchjourney.
Step2Dissectthebroadareaintosubareas.Attheonset,youwillrealisethatallthebroadareasmentionedabove–youthwelfare,refugees,domesticviolence,consumerbehaviourandHIV/AIDS–havemanyaspects.Forexample,therearemanyaspectsandissuesintheareaofdomesticviolence,illustratedinFigure4.1.
Similarly,youcanselectanysubjectareafromotherfieldssuchascommunityhealthorconsumerresearchandgothroughthisdissectionprocess.Inpreparingthislistofsubareasyoushouldalsoconsultotherswhohavesomeknowledgeoftheareaandtheliteratureinyoursubjectarea.Onceyouhavedevelopedanexhaustivelistofthesubareasfromvarioussources,youproceedtothenextstagewhereyouselectwhatwillbecomethebasisofyourenquiry.
Step3
Selectwhatisofmostinteresttoyou.Itisneitheradvisablenorfeasibletostudyallsubareas.Outofthislist,selectissuesorsubareasaboutwhichyouarepassionate.Thisisbecauseyourinterestshouldbethemostimportantdeterminantforselection,eventhoughthereareotherconsiderationswhichhavebeendiscussedintheprevioussection,‘Considerationsinselectingaresearchproblem’.Onewaytodecidewhatinterestsyoumostistostartwiththeprocessofelimination.Gothroughyourlistanddeleteallthosesubareasinwhichyouarenotveryinterested.Youwillfindthattowardstheendofthisprocess,itwillbecome very difficult for you to delete anything further. You need to continue until you are left with something that ismanageableconsideringthetimeavailabletoyou,yourlevelofexpertiseandotherresourcesneededtoundertakethestudy.Onceyouareconfidentthatyouhaveselectedanissueyouarepassionateaboutandcanmanage,youarereadytogotothenextstep.
Step4Raiseresearchquestions.Atthisstepaskyourself,‘WhatisitthatIwanttofindoutaboutinthissubarea?’Makealistofwhateverquestionscometoyourmindrelatingtoyourchosensubareaandifyouthinktherearetoomanytobemanageable,gothroughtheprocessofelimination,asyoudidinStep3.
Step5
Formulateobjectives.Bothyourmainobjectivesandyoursubobjectivesnowneedtobeformulated,whichgrowoutofyourresearch questions. Themain difference between objectives and research questions is the way in which they are written.Researchquestionsareobviouslythat–questions.Objectivestransformthesequestionsintobehaviouralaimsbyusingaction-orientedwordssuchas‘tofindout’, ‘todetermine’, ‘toascertain’and‘toexamine’.Someresearchersprefer toreverse theprocess; that is, theystart fromobjectivesandformulateresearchquestionsfromthem.Someresearchersaresatisfiedonlywithresearchquestions,anddonotformulateobjectivesatall.Ifyouprefertohaveonlyresearchquestionsoronlyobjectives,this is fine, but keep in mind the requirements of your institution for research proposals. For guidance on formulatingobjectives,seethelatersection.
Step6 Assessyourobjectives.Nowexamineyourobjectives toascertain the feasibilityof achieving them throughyour researchendeavour.Considertheminthelightofthetime,resources(financialandhuman)andtechnicalexpertiseatyourdisposal.
Step7
Double-check.Gobackandgivefinalconsiderationtowhetherornotyouaresufficientlyinterestedinthestudy,andhaveadequateresources toundertake it.Askyourself, ‘AmIreallyenthusiasticabout thisstudy?’and‘DoI reallyhaveenoughresources to undertake it?’ Answer these questions thoughtfully and realistically. If your answer to one of them is ‘no’,reassessyourobjectives.
Figures4.2to4.4operationaliseSteps1–7withexamplesfromdifferentacademicdisciplines(health,socialwork/socialsciencesandcommunitydevelopment).
Theformulationofresearchobjectives
Objectivesare thegoalsyousetout toattain inyourstudy.Since theseobjectives informareaderofwhat you want to achieve through the study, it is extremely important to word them clearly andspecifically.Objectivesshouldbelistedundertwoheadings:
mainobjectives;subobjectives.
Themainobjective isanoverall statementof the thrustofyour study. It isalsoa statementof themain associations and relationships that you seek to discover or establish. The subobjectives are thespecificaspectsofthetopicthatyouwanttoinvestigatewithinthemainframeworkofyourstudy.
Example1:Supposeyouwanttoconductastudyintheareaofalcoholism.Informulatingyourresearchproblemtakethefollowingsteps.
FIGURE4.2Stepsinformulatingaresearchproblem–alcoholism
Example2:Supposeyouwanttostudytherelationshipbetweenfertilityandmortality.Followthesesteps.
FIGURE4.3Formulatingaresearchproblem–therelationshipbetweenfertilityandmortality
Example3:Supposeyouwanttoconductastudyintheareaofhealth.Followthesesteps.
FIGURE4.4Narrowingaresearchproblem–health
Subobjectives should be numerically listed. They should be worded clearly and unambiguously.Makesurethateachsubobjectivecontainsonlyoneaspectof thestudy.Useaction-orientedwordsorverbswhenwritingyourobjectives.Theobjectivesshouldstartwithwordssuchas‘todetermine’,‘tofindout’,‘toascertain’,‘tomeasure’and‘toexplore’.The way the main objectives and subobjectives are worded determines how your research is
classified (e.g. descriptive, correlational or experimental). In other words, the wording of yourobjectivesdeterminesthetypeofresearchdesignyouneedtoadopttoachievethem.Hence,becarefulaboutthewayyouwordyourobjectives.Irrespectiveofthetypeofresearch,theobjectivesshouldbeexpressedinsuchawaythatthewording
clearly,completelyandspecificallycommunicatestoyourreadersyourintention.Thereisnoplaceforambiguity,non-specificityor incompleteness, either in thewordingofyourobjectivesor in the ideastheycommunicate.Figure4.5displaysthecharacteristicsofthewordingofobjectivesinrelationtothetypeofresearchstudy.
FIGURE4.5Characteristicsofobjectives
Ifyourstudyisprimarilydescriptive,yourmainobjectiveshouldclearlydescribethemajorfocusofyour study,evenmentioning theorganisationand its locationunless theseare tobekeptconfidential(e.g.todescribethetypesoftreatmentprogrammeprovidedby[nameoftheorganisation]toalcoholicsin[nameoftheplace]ortofindouttheopinionofthecommunityaboutthehealthservicesprovidedby[nameofthehealthcentre/department]in[nameoftheplace]).Identificationoftheorganisationanditslocation is important as the servicesmay be peculiar to the place and the organisation andmay notrepresenttheservicesprovidedbyotherstosimilarpopulations.Ifyourstudyiscorrelational innature, inadditionto thefirst threecharacteristicsshowninFigure
4.5,thewordingofthemainobjectiveshouldalsoincludethemainvariablesbeingcorrelated(e.g.toascertaintheimpactofmigrationonfamilyrolesor tocompare theeffectivenessofdifferent teachingmethodsonthecomprehensionofstudents).Iftheoverallthrustofyourstudyistotestahypothesis,thewordingofthemainobjectivesshould
also indicate the direction of the relationship being tested (e.g. to ascertain if an increase in youthunemployment will increase the incidence of street crime, or to demonstrate that the provision ofmaternalandchildhealthservicestoAboriginalpeopleinruralAustraliawillreduceinfantmortality).
Thestudypopulation
Sofarwehavefocusedononlyoneaspectofastudy,theresearchproblem.Buteverystudyinsocialsciences has a second aspect, the study population, from whom the required information to findanswerstoyourresearchquestionsisobtained.Asyounarrowtheresearchproblem,similarlyyouneedto decide very specifically and clearly who constitutes your study population, in order to select theappropriaterespondents.Supposeyouhavedesignedastudytoascertaintheneedsofyoungpeoplelivinginacommunity.In
termsofthestudypopulation,oneofthefirstquestionsyouneedtoansweris:‘WhodoIconsidertobeayoungperson?’Youneedtodecide, inmeasurableterms,whichagegroupyourrespondentsshouldcome from. Is it those between 15 and 18, 15 and 20 or 15 and 25 years of age? Or you may beinterestedinsomeotheragegroup.Youneedtodecidethisbeforeundertakingyourresearchjourney.Having decided the age group that constitutes your ‘young person’, the next question you need toconsideriswhetheryouwanttoselectyoungpeopleofeithergenderorconfinethestudytooneonly.Inaddition, there is another dimension to consider: that is, what constitutes the community? Whichgeographicalarea(s)orethnicbackgroundshouldIselectmyrespondentsfrom?Letustakeanotherexample.Supposeyouwanttofindoutthesettlementprocessofimmigrants.Asa
partofidentifyingyourstudypopulation,youneedtodecidewhowouldyouconsideranimmigrant.Isit apersonwho immigrated5,10,15or20yearsago?Youalsoneed toconsider thecountries fromwhere the immigrantscome.Willyouselectyourrespondents irrespectiveof thecountryoforiginor
select only those who have come from a specific country(ies)? In a way you need to narrow yourdefinition of the study population as you have done with your research problem. These issues arediscussedingreaterdepthunder‘Establishingoperationaldefinitions’followingthissection.Inquantitativeresearch,youneedtonarrowboththeresearchproblemandthestudypopulationand
make them as specific as possible so that you and your readers are clear about them. In qualitativeresearch,reflectingthe‘exploratory’philosophicalbaseoftheapproach,boththestudypopulationandtheresearchproblemshouldremainlooseandflexibletoensurethefreedomnecessarytoobtainvariedandrichdataifasituationemerges.
Establishingoperationaldefinitions
In defining the problemyoumay use certainwords or items that are difficult tomeasure and/or theunderstandingofwhichmayvaryfromrespondenttorespondent.Inaresearchstudyitisimportanttodevelop, define or establish a set of rules, indicators or yardsticks in order to establish clearly themeaningofsuchwords/items.Itissometimesalsoimportanttodefineclearlythestudypopulationfromwhich you need to obtain the required information.When you define concepts that you plan to useeither inyourresearchproblemand/or in identifyingthestudypopulationinameasurableform, theyare calledworking definitions oroperational definitions. Youmust understand that theseworkingdefinitionsthatyoudevelopareonlyforthepurposeofyourstudyandcouldbequitedifferenttolegaldefinitions,orthoseusedbyothers.Astheunderstandingofconceptscanvarymarkedlyfrompersontoperson,yourworkingdefinitionswillinformyourreaderswhatexactlyyoumeanbytheconceptsthatyouhaveusedinyourstudy.Thefollowingexamplestudieshelptoexplainthis.Themainobjectivesare:
1. TofindoutthenumberofchildrenlivingbelowthepovertylineinAustralia.2. Toascertaintheimpactofimmigrationonfamilyrolesamongimmigrants.3. Tomeasuretheeffectivenessofaretrainingprogrammedesignedtohelpyoungpeople.
Althoughtheseobjectivesclearlystatethemainthrustofthestudies,theyarenotspecificintermsofthemainvariables tobestudiedand thestudypopulations.Youcannotcount thenumberofchildrenlivingbelowthepovertylineuntilyoudecidewhatconstitutesthepovertylineandhowtodetermineit;youcannotfindouttheimpactofimmigrationonfamilyrolesunlessyouidentifywhichrolesconstitutefamilyroles;andyoucannotmeasureeffectivenessuntilyoudefinewhateffectivenessis.Ontheotherhand,it isequallyimportanttodecideexactlywhatyoumeanby‘children’,‘immigrants’or‘young’.Uptowhatagewillyouconsiderapersontobeachild(i.e.5,10,15or18)?Whowouldyouconsideryoung?Aperson15yearsofage,20,25or30?Whowouldyouconsidertobeanimmigrant?Apersonwhoimmigrated40,20or5yearsago?Inaddition,areyougoingtoconsiderimmigrantsfromeverycountryoronlyafew?Inmanycasesyouneedtodevelopoperationaldefinitionsforthevariablesandconceptsyouarestudyingandforthepopulationthatbecomesthesourceoftheinformationforyourstudy. Table 4.2 lists the concepts and the population groups to be operationalised for the aboveexamples.
TABLE4.2Operationalisationofconceptsandthestudypopulations
Inaresearchstudyyouneedtodefinetheseclearlyinordertoavoidambiguityandconfusion.Thisisachieved through the process of developing operational/working definitions. You need to developoperationaldefinitionsforthemajorconceptsyouareusinginyourstudyanddevelopaframeworkforthestudypopulationenablingyoutoselectappropriaterespondents.Operational definitions may differ from day-to-day meanings as well as dictionary or legal
definitions. These meanings may not be helpful in identifying either your study population or theconcepts you are studying.Though in daily life you often usewords such as ‘children’, ‘youth’ and‘immigrant’ loosely, you need to bemore specificwhen using them in a research study.You shouldworkthroughyourowndefinitions.Operationaldefinitionsgiveanoperationalmeaningtothestudypopulationandtheconceptsused.It
isonlythroughmakingyourproceduresexplicitthatyoucanvalidlydescribe,explain,verifyandtest.Itisimportanttorememberthattherearenorulesfordecidingifanoperationaldefinitionisvalid.Yourargumentsmustconvinceothersabouttheappropriatenessofyourdefinitions.
Formulatingaresearchprobleminqualitativeresearch
Thedifference inqualitativeandquantitativestudiesstartswith thewayyouformulateyour researchproblem. In quantitative research you strive to be as specific as possible, attempt to narrow themagnitudeofyourstudyanddevelopaframeworkwithinwhichyouconfineyoursearch.Ontheotherhand, in qualitative research, this specificity in scope,methods and framework is almost completelyignored.Youstrivetomaintainflexibility,opennessandfreedomtoincludeanynewideasorexcludeanyaspectthatyouinitiallyincludedbutlaterconsidernottoberelevant.Attheinitialstageyouonlyidentify themain thrust of your study and some specific aspectswhich youwant to find out about.Qualitativeresearchprimarilyemploysinductivereasoning.Incontrasttoquantitativeresearch,wherearesearch problem is stated before data collection, in qualitative research the problem is reformulatedseveraltimesafteryouhavebegunthedatacollection.Theresearchproblemaswellasdatacollectionstrategiesare reformulatedasnecessary throughoutdatacollectioneither toacquire the ‘totality’ofaphenomenonortoselectcertainaspectsforgreaterin-depthstudy.Thisflexibilityandfreedom,thoughprovidingyouwithcertainadvantages,canalsocreateproblems
in terms of comparability of the information gathered. It is possible that your areas of search maybecomemarkedlydifferentduringthepreliminaryandfinalstagesofdatagathering.Duringtheinitialdevelopmentalphase,manyresearchersproduceaframeworkof‘reminders’(aconceptualframeworkofenquiry)toensurethatkeyissues/aspectsarecoveredduringdiscussionswiththerespondents.Asthestudy progresses, if needs be, issues or themes are added to this framework. This is not a list ofquestions but reminders that are only used if for some reason the interactionwith respondents lacksdiscussion.Let us take an example to detail the process of formulation of a research problem in qualitative
research:
OnceIsupervisedastudentwhowasinterestedinattention-deficithyperactivitydisorder(ADHD).Shewanted to find out, as she put it, ‘What does itmeans to have a childwithADHD in thefamily?’Ofcoursemyfirstquestiontoherwas,‘Whatdoyoumeanby“whatdoesitmean”?’Shepausedforawhileandthensaid,‘itmeanswhatitmeans’.Iaskedhertotreatmeasoneofherrespondentsandaskthequestion.Sheaskedme,‘WhatdoesitmeantohaveachildwithADHD?’towhichmyanswerwas,‘Idonotunderstandyourquestion.Couldyoupleaseexplaintomethemeaningof“whatdoes itmean”?’She found itdifficult toexplainand immediately realised theproblem with the question. What she thought was very clear to her became quite difficult toexplain. It took her awhile to explain tomewhat she had inmind.During the discussion thatfollowed,thoughshecouldexplainsomeofthethingsshehadinmind,sherealisedthatshecouldnotgotoarespondentwithherinitialquestion.ThestudentknewafamilywhohadachildwithADHDfromwhichherinterestinthetopichadprobablystemmed.Isuggested
thatshehaveatalkwiththemother.Shedid,and,tohersurprise,themotheraskedherthesamequestionthatIhad.IadvisedhertoreadsomeliteratureonADHDandalsohaveinformaltalkswithtwofamilieswhohaveachildwithADHD.We
decidedtoselectonesinglemotherfamilyandtheotherwherethefatherandthemotherbothtakeresponsibilityforthechild.Shewasadvisedtorecordall theissuesandaspectsthatreflectedherunderstandingof‘whatdoesitmean’,relatingtobringingupachildwithADHDinthefamily.Aftergoingthroughtheabove,shedevelopedalistthreeandahalfpageslongoftheaspectsandissuesthat,accordingtoher,reflectedherunderstandingof‘whatdoesitmean’.Shedidnotconstructanyspecificquestionsaroundtheseaspectsorissues.Theyservedasbackgroundforhertoraisewithpotentialrespondentsincaserespondentsdidnotcomeupwithissuesoraspectsfordiscussionintermsof‘WhatdoesitmeantohaveachildwithADHDinthefamily?’Thislistbroughtimmenseclarificationtoherthinkingabout‘whatdoesitmean’andservedasthebasisofherinterviewswith
thefamilies.Anumberoftimesduringthesupervisorysessionsshehadmentionedthatshewouldnothavebeenabletodomuchwithouttheconceptualframework.Youshouldnotconfuseitwiththeinterviewguide.Thelistisaconceptualconstructionofthethoughts that serve asbackgroundandbecome thebasisofdiscussions in case there is insufficientdialoguewithyourpotentialrespondents.
SummaryTheformulationofaresearchproblemisthemostimportantstepintheresearchprocess.Itisthefoundation,intermsofdesign,onwhichyoubuildthewholestudy.Anydefectsinitwilladverselyaffectthevalidityandreliabilityofyourstudy.Therearenospecificguidelinesbutthemodelsuggestedinthischaptercouldserveasausefulframeworkforthebeginner.The
seven-stepmodelhelpsyoutonarrowyourbroadareaofinteresttoenableyoutodecidewhatspecificallyyouwanttostudy.Itisoperational in nature and follows a logical sequence that takes the beginner through the complexities of formulating a researchprobleminasimpleandeasy-to-understandmanner.Itisimportanttoarticulatetheobjectivesofyourstudyclearly.Objectivesshouldbespecificandfreefromambiguity,andeach
oneshouldrelatetoonlyoneaspectofthestudy.Theyshouldbeundertwoheadings:mainobjectiveandsubobjectives.Useaction-orientedwordswhenwritingyourobjectives.Formulation of a research problem in qualitative research follows a different path.You do not
predetermine theexactnatureandextentof theresearchproblemyoupropose to findanswers to.Youcontinuetomodifyitasyoustartfindingoutmoreaboutit.However, itwillhelpyouifyoudevelop a conceptual frameworkof thedifferent aspects of aproblem to serve as abackdrop forissuestobediscussedwithpotentialrespondents.Developingoperationaldefinitionsfortheconceptsthatyouproposetostudyisextremelyimportant.Thisenhancesclarityabout
theissuesyouaretryingtofindoutaboutandaboutthestudypopulationyouplantogatherinformationfrom.Itisimportantthatyouoperationaliseboththemainvariablesyouareproposingtostudyandthestudypopulation.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Identifytwoorthreepotentialresearchquestions,relatedtoyourownacademicfieldorprofessionalarea,thatwouldfallundereachofthefourPs(asoutlinedinTable4.1):
people;problems;programs;phenomena.
Foreachofthesehypotheticalresearchquestions,identifywhichconceptsandstudypopulationswouldneedtobeoperationallydefined.Considerwhatproblemsmightoccurifthiswasnotdone.Selectabroadsubjectareaofinteresttoyouand‘dissect’itintosubareas.
CHAPTER5IdentifyingVariables
Inthischapteryouwilllearnabout:
WhatvariablesandconceptsareandhowtheyaredifferentHowtoturnconceptsintooperationalvariablesTypesofvariablesfromtheviewpointof:
CausationThestudydesignTheunitofmeasurement
Typesofmeasurementscales:
ThenominalorclassificatoryscaleTheordinalorrankingscaleTheintervalscaleTheratioscale
Keywords: active variables, attribute variables, categorical variables, causation,constant variables, continuous variables, dependent variables, dichotomous,extraneous variables, independent variables, interval scale, intervening variables,measurement scales, nominal scale, ordinal scale, polytomous, ratio scale, unit ofmeasurement.
Ifitexists,itcanbemeasured.(Babbie1989:105)
In the process of formulating a research problem, in the case of quantitative research, there are twoimportant considerations: the use of concepts and the construction of hypotheses. In the previouschapter,we established that concepts are highly subjective as an understanding of them varies frompersontoperson.Itfollows,therefore,thatassuchtheymaynotbemeasurable.Inaresearchstudyitisimportant that the concepts used should be operationalised inmeasurable terms so that the extent ofvariationinrespondents’understandingisreducedifnoteliminated.Usingtechniquestooperationalise
concepts,andknowledgeaboutvariables,playsanimportantroleinreducingthisvariabilityand‘finetuning’yourresearchproblem.
Whatisavariable?
Whetherweacceptitornot,weallmakevaluejudgementsconstantlyinourdailylives:‘Thisfoodisexcellent’;‘Icouldnotsleepwelllastnight’;‘Idonotlikethis’;and‘Ithinkthisiswonderful’.Theseare all judgements based upon ourown preferences, indicators or assessment.Because these explainfeelingsorpreferences, thebasisonwhich theyaremademayvarymarkedlyfromperson toperson.Thereisnouniformyardstickwithwhichtomeasurethem.Aparticularfoodmaybejudged‘excellent’byonepersonbut‘awful’byanother,andsomethingelsecouldbewonderfultoonepersonbutuglytoanother.Whenpeopleexpressthesefeelingsorpreferences,theydosoonthebasisofcertaincriteriaintheirminds,orinrelationtotheirexpectations.Ifyouweretoquestionthemyouwilldiscoverthattheirjudgement is based upon indicators and/or expectations that lead them to conclude and express aparticularopinion.Letusconsiderthisinaprofessionalcontext:
‘Thisprogrammeiseffective.’‘Thisprogrammeisnoteffective.’‘Weareprovidingaqualityservicetoourclients.’‘Thisisawasteoftime.’‘Inthisinstitutionwomenarediscriminatedagainst.’‘Thereisnoaccountabilityinthisoffice.’‘Thisproductisnotdoingwell.’
These are not preferences per se; these are judgements that require a sound basis on which toproclaim.Forexample,ifyouwanttofindoutifaprogrammeiseffective,ifaserviceisofqualityorifthere isdiscrimination,youneed tobecareful that such judgementshavea rationaland soundbasis.Thiswarrantstheuseofameasuringmechanismanditisintheprocessofmeasurementthatknowledgeaboutvariablesplaysanimportantrole.An image, perception or concept that is capable of measurement – hence capable of taking on
different values – is called a variable. In other words, a concept that can be measured is called avariable. According to Kerlinger, ‘A variable is a property that takes on different values. Putting itredundantly,avariableissomethingthatvaries…Avariableisasymboltowhichnumeralsorvaluesareattached’(1986:27).BlackandChampiondefineavariableas‘rationalunitsofanalysis thatcanassumeanyoneofanumberofdesignatedsetsofvalues’(1976:34).Aconceptthatcanbemeasuredonany one of the four types of measurement scale, which have varying degrees of precision inmeasurement,iscalledavariable(measurementscalesarediscussedlaterinthischapter).However, there are somewhobelieve that scientificmethods are incapable ofmeasuring feelings,
preferences, values and sentiments. In the author’s opinion most of these things can be measured,thoughtherearesituationswheresuchfeelingsorjudgementscannotbedirectlymeasuredbutcanbemeasured indirectly through appropriate indicators. These feelings and judgements are based uponobservablebehaviours in real life, though theextent towhich thebehaviours reflect their judgementsmayvaryfrompersontoperson.CohenandNagelexpresstheiropinioninthefollowingwords:
There are, indeed, a great many writers who believe that scientific method is inherentlyinapplicable to such judgements as estimation or value, as ‘This is beautiful’, ‘This is good’ or‘Thisoughttobedone’…alljudgementsofthelattertypeexpressnothingbutfeelings,tastesorindividualpreferences,suchjudgementscannotbesaidtobetrueorfalse(exceptasdescriptionsofthepersonal feelingsof theonewhoutters them)…Almostallhumandiscoursewouldbecomemeaninglessifwetooktheviewthateverymoraloraestheticjudgementisnomoretrueorfalsethananyother.(1966:352)
Thedifferencebetweenaconceptandavariable
Measurabilityisthemaindifferencebetweenaconceptandavariable.Conceptsarementalimagesorperceptions and therefore their meanings vary markedly from individual to individual, whereasvariables aremeasurable, though, of course,with varying degrees of accuracy.A concept cannot bemeasuredwhereasavariablecanbesubjectedtomeasurementbycrude/refinedorsubjective/objectiveunits ofmeasurement. Concepts are subjective impressionswhich, ifmeasured as suchwould causeproblemsincomparingresponsesobtainedfromdifferentrespondents.AccordingtoYoung:
Eachcollaboratormusthavethesameunderstandingoftheconceptsifthecollaborativedataaretobe similarly classified and the findings pooled and tested, or reproduced. Classification andcomparison demand uniform and precise definitions of categories expressed in concepts. (1966:18)
Itisthereforeimportantfortheconceptstobeconvertedintovariables(eitherdirectlyorthroughasetof indicators)as theycanbesubjected tomeasurement,even though thedegreeofprecisionwithwhichtheycanbemeasuredmarkedlyvariesfromonemeasurementscaletoanother(nominal,ordinal,interval and ratio). Table 5.1 gives examples of concepts and variables to illustrate the differencesbetweenthem.
TABLE5.1ExamplesofconceptsandvariablesConcepts Variables
EffectivenessSatisfactionImpactExcellentHighachieverSelf-esteemRichDomesticviolenceExtentandpatternofalcoholconsumptionetc.
Gender(male/female)AttitudeAge(xyears,ymonths)Income($__peryear)Weight(__kg)Height(__cm)Religion(Catholic,protestant,Jew,Muslim)etc.
SubjectiveimpressionNouniformityastoitsunderstandingamong
Measurablethoughthedegreeofprecisionvariesfromscaletoscaleandfromvariable
differentpeopleAssuchcannotbemeasured
tovariable(e.g.attitude–subjective,income–objective)
Convertingconceptsintovariables
Ifyouareusingaconceptinyourstudy,youneedtoconsideritsoperationalisation–thatis,howitwillbemeasured. Inmost cases, to operationalise a concept you first need to go through the process ofidentifying indicators– a set of criteria reflectiveof the concept–which can thenbe converted intovariables.Thechoiceofindicatorsforaconceptmightvarywiththeresearcherbutthoseselectedmusthavealogicallinkwiththeconcept.Someconcepts,suchas‘rich’(intermsofwealth),caneasilybeconvertedintoindicatorsandthenvariables.Forexample,todecideobjectivelyifapersonis‘rich’,onefirstneedstodecideupontheindicatorsofwealth.Assumethatwedecideuponincomeandassetsastheindicators.Incomeisalsoavariablesinceitcanbemeasuredindollars;therefore,youdonotneedto convert this into a variable.Although the assets owned by an individual are indicators of his/her‘richness’, theystillbelong to thecategoryofconcepts.Youneed to look furtherat the indicatorsofassets.Forexample,house,boat,carandinvestmentsareindicatorsofassets.Convertingthevalueofeachoneintodollarswillgivethetotalvalueoftheassetsownedbyaperson.Next,fixalevel,baseduponavailableinformationonincomedistributionandanaveragelevelofassetsownedbymembersofacommunity,whichactsasthebasisforclassification.Thenanalysetheinformationonincomeandthetotalvalueoftheassetstomakeadecisionaboutwhetherthepersonshouldbeclassifiedas‘rich’.Theoperationalisationofotherconcepts,suchasthe‘effectiveness’or‘impact’ofaprogramme,mayprovemore difficult. Table 5.2 shows some examples that will help you to understand the process ofconvertingconceptsintovariables.Oneof themaindifferences betweenquantitative andqualitative research studies is in the area of
variables.Inqualitativeresearch,asitusuallyinvolvesstudyingperceptions,beliefs,orfeelings,youdonotmakeanyattempttoestablishuniformityinthemacrossrespondentsandhencemeasurementsandvariablesdonotcarrymuchsignificance.Ontheotherhand,inquantitativestudies,astheemphasisisonexploringcommonalitiesinthestudypopulation,measurementsandvariablesplayanimportantrole.
TABLE5.2Convertingconceptsintovariables
Typesofvariable
Avariablecanbeclassifiedinanumberofways.Theclassificationdevelopedhereresultsfromlookingatvariablesinthreedifferentways(seeFigure5.1):
thecausalrelationship;thestudydesign;theunitofmeasurement.
Fromtheviewpointofcausalrelationship
In studies that attempt to investigate a causal relationship or association, four sets of variablesmayoperate(seeFigure5.2):
1. changevariables,whichareresponsibleforbringingaboutchangeinaphenomenon,situationorcircumstance;
2. outcomevariables,whicharetheeffects,impactsorconsequencesofachangevariable;
3. variableswhichaffectorinfluencethelinkbetweencause-and-effectvariables;4. connectingorlinkingvariables,whichincertainsituationsarenecessarytocompletethe
relationshipbetweencause-and-effectvariables.
In research terminology,changevariablesarecalled independentvariables,outcome/effectvariablesare calleddependentvariables, the unmeasured variables affecting the cause-and-effect relationshipare calledextraneousvariables and the variables that link a cause-and-effect relationship are calledinterveningvariables.Hence:
1. Independentvariable–thecausesupposedtoberesponsibleforbringingaboutchange(s)inaphenomenonorsituation.
2. Dependentvariable–theoutcomeorchange(s)broughtaboutbyintroductionofanindependentvariable.
3. Extraneousvariable–severalotherfactorsoperatinginareal-lifesituationmayaffectchangesinthedependentvariable.Thesefactors,notmeasuredinthestudy,mayincreaseordecreasethemagnitudeorstrengthoftherelationshipbetweenindependentanddependentvariables.
4. Interveningvariable–sometimescalledtheconfoundingvariable(Grinnell1988:203),itlinkstheindependentanddependentvariables.Incertainsituationstherelationshipbetweenanindependentandadependentvariablecannotbeestablishedwithouttheinterventionofanothervariable.Thecause,orindependent,variablewillhavetheassumedeffectonlyinthepresenceofaninterveningvariable.
FIGURE5.1Typesofvariable
Note:Classificationacrossaclassificationbaseisnotmutuallyexclusivebutclassificationwithinaclassificationbaseis.Withinastudyanindependentvariablecanbeanactivevariable,oraquantitativeoraqualitativevariableanditcanalsobeacontinuousoracategoricalvariablebutitcannotbeadependent,anextraneousoraninterveningvariable.
FIGURE5.2Typesofvariableinacausalrelationship
Toexplainthesevariablesletusconsidersomeexamples.Supposeyouwanttostudytherelationshipbetweensmokingandcancer.Youassumethatsmokingisacauseofcancer.Studieshaveshownthatthere aremany factors affecting this relationship, such as the number of cigarettes or the amount oftobacco smoked every day; the duration of smoking; the age of the smoker; dietary habits; and theamountof exerciseundertakenby the individual.All of these factorsmayaffect the extent towhich
smoking might cause cancer. These variables may either increase or decrease the magnitude of therelationship.In the above example the extent of smoking is the independent variable, cancer is the dependent
variable and all the variables that might affect this relationship, either positively or negatively, areextraneousvariables.SeeFigure5.3.
FIGURE5.3Independent,dependentandextraneousvariablesinacausalrelationship
Letustakeanotherexample.Supposeyouwanttostudytheeffectsofamarriagecounsellingserviceonmaritalproblemsamongclientsofanagencyprovidingsuchaservice.Figure5.4showsthesetsofvariablesthatmayoperateinstudyingtherelationshipbetweencounsellingandmarriageproblems.
FIGURE5.4Setsofvariablesincounsellingandmarriageproblems
Instudyingtherelationshipbetweenacounsellingserviceandmarriageproblems,itisassumedthatthecounsellingservicewillinfluencetheextentofmaritalproblems.Hence,inthestudyoftheabove
relationship, the type of counselling service is the independent variable and the extent of marriageproblems is the dependent variable. The magnitude or strength of this relationship can be affected,positively or negatively, by a number of other factors that are not the focus of the study. Theseextraneousvariablesmightbe thebirthofachild; improvement inacouple’seconomicsituation; thecouple’smotivation to change the situation; the involvement of another person; self-realisation; andpressurefromrelativesandfriends.Extraneousvariablesthatworkbothwayscanincreaseordecreasethestrengthoftherelationship.TheexampleinFigure5.5shouldhelpyoutounderstandinterveningvariables.Supposeyouwantto
studytherelationshipbetweenfertilityandmortality.Youraimis toexplorewhathappenstofertilitywhen mortality declines. The history of demographic transition has shown that a reduction in thefertilitylevelfollowsadeclineinthemortalitylevel,thoughthetimetakentoattainthesamelevelofreductioninfertilityvariedmarkedlyfromcountry tocountry.Assuch, there isnodirectrelationshipbetween fertility and mortality. With the reduction in mortality, fertility will decline only if peopleattempttolimittheirfamilysize.Historyhasshownthatforamultiplicityofreasons(thediscussionofwhich is beyond the scope of this book) people have used one method or another to control theirfertility, resulting in lower fertility levels. It is thus the intervention of contraceptive methods thatcompletestherelationship:thegreatertheuseofcontraceptives,thegreaterthedeclineinthefertilitylevelandthesoonertheadoptionofcontraceptivemethodsbypeople,thesoonerthedecline.Theextentoftheuseofcontraceptivesisalsoaffectedbyanumberofotherfactors,forexampleattitudestowardscontraception,levelofeducation,socioeconomicstatusandage,religion,andprovisionandqualityofhealthservices.Theseareclassifiedasextraneousvariables.
FIGURE5.5Independent,dependent,extraneousandinterveningvariables
Intheaboveexample,declineinmortalityisassumedtobethecauseofareductioninfertility,hencethe mortality level is the independent variable and fertility is the dependent variable. But thisrelationshipwillbecompletedonlyifanothervariableintervenes–thatis,theuseofcontraceptives.Areductioninmortality(especiallychildmortality)increasesfamilysize,andanincreaseinfamilysizecreates a number of social, economic and psychological pressures on families, which in turn createattitudes favourable toa smaller family size.Thischange inattitudes iseventuallyoperationalised inbehaviour through theadoptionofcontraceptives. Ifpeopledonotadoptmethodsofcontraception,achange in mortality levels will not be reflected in fertility levels. The population explosion in
developingcountriesisprimarilyduetolackofacceptanceofcontraceptives.Theextentoftheuseofcontraceptivesdeterminesthelevelofthedeclineinfertility.Theextentofcontraceptiveadoptionbyapopulation is dependent upon a number of factors. As mentioned earlier, in this causal model, thefertilitylevelis thedependentvariable, theextentofcontraceptiveuseis theinterveningvariable, themortalitylevel is theindependentvariable,andtheunmeasuredvariablessuchasattitudes,education,age, religion, thequalityof services, and soon, are all extraneousvariables.Without the interveningvariabletherelationshipbetweentheindependentanddependentvariableswillnotbecomplete.
FIGURE5.6Activeandattributevariables
Fromtheviewpointofthestudydesign
A study that examines association or causation may be a controlled/contrived experiment, a quasi-experimentor anexpost facto or non-experimental study. In controlled experiments the independent(cause)variablemaybeintroducedormanipulatedeitherbytheresearcherorbysomeoneelsewhoisprovidingtheservice.Inthesesituationstherearetwosetsofvariables(seeFigure5.6):
Activevariables–thosevariablesthatcanbemanipulated,changedorcontrolled.Attributevariables–thosevariablesthatcannotbemanipulated,changedorcontrolled,andthatreflectthecharacteristicsofthestudypopulation,forexampleage,gender,educationandincome.
Supposeastudyisdesignedtomeasuretherelativeeffectivenessofthreeteachingmodels(ModelA,ModelBandModelC).Thestructureandcontentsofthesemodelscouldvaryandanymodelmightbetestedonanypopulationgroup.Thecontents,structureandtestabilityofamodelonapopulationgroupmayalsovaryfromresearchertoresearcher.Ontheotherhand,aresearcherdoesnothaveanycontrolover characteristicsof the studentpopulation suchas their age, genderormotivation to study.Thesecharacteristicsof thestudypopulationarecalledattributevariables.However,a researcherdoeshavetheability tocontrol and/orchange the teachingmodels.S/hecandecidewhatconstitutesa teachingmodelandonwhichgroupofthestudentpopulationitshouldbetested(ifrandomisationisnotused).
Fromtheviewpointoftheunitofmeasurement
Fromtheviewpointoftheunitofmeasurement,therearetwowaysofcategorisingvariables:
whethertheunitofmeasurementiscategorical(asinnominalandordinalscales)orcontinuousinnature(asinintervalandratioscales);
whetheritisqualitative(asinnominalandordinalscales)orquantitativeinnature(asinintervalandratioscales).
Onthewholethereisverylittledifferencebetweencategoricalandqualitative,andbetweencontinuousandquantitative,variables.Theslightdifferencebetweenthemisexplainedbelow.Categorical variables are measured on nominal or ordinal measurement scales, whereas for
continuousvariablesthemeasurementsaremadeoneitheranintervaloraratioscale.Therearethreetypesofcategoricalvariables:
constantvariable–hasonlyonecategoryorvalue,forexampletaxi,treeandwater;dichotomousvariable–hasonlytwocategories,asinmale/female,yes/no,good/bad,head/tail,up/downandrich/poor;polytomousvariable–canbedividedintomorethantwocategories,forexamplereligion(Christian,Muslim,Hindu);politicalparties(Labor,Liberal,Democrat);andattitudes(stronglyfavourable,favourable,uncertain,unfavourable,stronglyunfavourable).
Continuous variables, on the other hand, have continuity in their measurement, for example age,incomeandattitudescore.Theycantakeanyvalueonthescaleonwhichtheyaremeasured.Agecanbemeasuredinyears,monthsanddays.Similarly,incomecanbemeasuredindollarsandcents.Inmanywaysqualitativevariablesaresimilartocategoricalvariablesasbothuseeithernominalor
ordinalmeasurementscales.However,therearesomedifferences.Forexample,itispossibletodevelopcategoriesonthebasisofmeasurementsmadeonacontinuousscale,suchasmeasuringtheincomeofapopulation in dollars and cents and then developing categories such as ‘low’, ‘middle’ and ‘high’income. The measurement of income in dollars and cents is classified as the measurement of acontinuousvariable,whereasitssubjectivemeasurementincategoriessuchas‘low’,‘middle’and‘high’groupsisaqualitativevariable.Although this distinction exists, for most practical purposes there is no real difference between
categoricalandqualitativevariablesorbetweencontinuousandquantitativevariables.Table5.3showssimilaritiesanddifferencesamongthevarioustypesofvariable.
TABLE5.3Categorical/continuousandquantitative/qualitativevariables
*Canbeclassifiedinqualitativecategories,e.g.old,young,child;orquantitativelyonacontinuousscale,e.g.inyears,monthsanddays.^Canbemeasuredquantitativelyindollarsandcentsaswellasqualitativelyincategoriessuchashigh,middleandlow.+similarly,temperaturecanbemeasuredquantitativelyindegreesondifferentscales(Celsius,Fahrenheit)orinqualitativecategoriessuchashotandcold.
Forabeginneritisimportanttounderstandthatthewayavariableismeasureddeterminesthetypeofanalysisthatcanbeperformed,thestatisticalproceduresthatcanbeappliedtothedata,thewaythedatacanbeinterpretedandthefindingsthatcanbecommunicated.Youmaynotrealiseinthebeginningthat the style of your report is entirely dependent upon the way the different variables have beenmeasured–thatis,thewayaquestionhasbeenaskedanditsresponserecorded.Thewayyoumeasurethevariables inyourstudydetermineswhetherastudy is ‘qualitative’or ‘quantitative’ innature. It isthereforeimportanttoknowaboutthemeasurementscalesforvariables.
Typesofmeasurementscale
Theframeintowhichwewishtomakeeverythingfitisoneofourownconstruction;butwedonotconstructitatrandom,weconstructitbymeasurementsotospeak;andthatiswhywecanfitthefactsintoitwithoutalteringtheiressentialqualities.(Poincaré1952:xxv)
Measurementiscentral toanyenquiry.Inadditiontotheideologyandphilosophythatunderpineachmodeofenquiry,themostsignificantdifferencebetweenqualitativeandquantitativeresearchstudiesisinthetypesofmeasurementusedincollectinginformationfromtherespondents.Qualitativeresearchmostlyusesdescriptive statements to seek answers to the researchquestions,whereas inquantitativeresearchtheseanswersareusuallysoughtononeofthemeasurementscales(nominal,ordinal,intervalorratio).Ifapieceofinformationisnotcollectedusingoneofthescalesatthetimeofdatacollection,itistransformedintovariablesbyusingthesemeasurementscalesatthetimeofanalysis.Measurementon these scales could be either in the form of qualitative categories or through a precise unit ofmeasurement.Thosescaleswhichhaveaunitofmeasurement(intervalandratio)areconsideredtobemore refined, objective and accurate. On the other hand, nominal and ordinal scales are consideredsubjectiveandhencenotasaccurateastheydonothaveaunitofmeasurementperse.Thegreatertherefinementintheunitofmeasurementofavariable,thegreatertheconfidenceplacedinthefindingsbyothers, other things being equal. One of the main differences between the physical and the socialsciences is the units of measurement used and the degree of importance attached to them. In thephysical sciences measurements have to be absolutely accurate and precise, whereas in the socialsciencestheymayvaryfromtheverysubjectivetotheveryquantifiable.Withinthesocialsciencestheemphasis on precision in measurement varies markedly from one discipline to another. Ananthropologist normally uses very ‘subjective’ units of measurement, whereas an economist or anepidemiologistemphasises‘objective’measurement.There are two main classification systems in the social sciences for measuring different types of
variable.OnewasdevelopedbyS.S.Stevens(in1946)andtheotherbyDuncan(in1984).AccordingtoSmith(1991:72),‘Duncan(1984)hasenumerated, inincreasingorderofinterest toscientists,fivetypes of measurement: nominal classification, ordinal scaling, cardinal scaling, ratio scaling, andprobabilityscaling’.DuncanwritesaboutStevens’sclassificationasfollows:
Thetheoryofscaletypesproposedin1946bySSStevensfocusedonnominal,ordinal,interval,andratioscalesofmeasurement.Someofhisexamplesofthesetypes–notablythoseconcerningpsychologicaltestscores–aremisleading.(1984:viii)
However,Baileyconsidersthat‘SSStevensconstructedawidelyadoptedclassificationoflevelsofmeasurement’ (1978: 52). As this book iswritten for the beginner and as Stevens’s classification issimpler,itisthisthatisusedfordiscussioninthischapter.Stevenshasclassifiedthedifferenttypesof
measurementscaleintofourcategories:
nominalorclassificatoryscale;ordinalorrankingscale;intervalscale;ratioscale.
Table5.4summarisesthecharacteristicsofthefourscales.
TABLE5.4Characteristicsandexamplesofthefourmeasurementscales
Thenominalorclassificatoryscale
A nominal scale enables the classification of individuals, objects or responses based on acommon/sharedpropertyorcharacteristic.Thesepeople,objectsorresponsesaredividedintoanumberofsubgroupsinsuchawaythateachmemberofthesubgrouphasacommoncharacteristic.Avariablemeasuredonanominalscalemayhaveone, twoormoresubcategoriesdependingupontheextentofvariation.Forexample,‘water’and‘taxi’haveonlyonesubgroup,whereasthevariable‘gender’canbeclassified into two subcategories: male and female. Political parties in Australia can similarly beclassified into four main subcategories: Labor, Liberal, Democrats and Greens. Those who identifythemselves,eitherbymembershiporbelief,asbelongingtotheLaborPartyareclassifiedas‘Labor’,
those identifying with the Liberals are classified as ‘Liberal’, and so on. The name chosen for asubcategory isnotional,but foreffectivecommunication it isbest tochoosesomething thatdescribesthecharacteristicofthesubcategory.Classificationbymeansofanominalscaleensuresthatindividuals,objectsorresponseswithinthe
samesubgrouphaveacommoncharacteristicorpropertyasthebasisofclassification.Thesequenceinwhichsubgroupsarelistedmakesnodifferenceasthereisnorelationshipamongsubgroups.
Theordinalorrankingscale
Anordinalscalehasallthepropertiesofanominalscale–categorisingindividuals,objects,responsesorapropertyintosubgroupsonthebasisofacommoncharacteristic–butalsoranksthesubgroupsinacertainorder.Theyarearrangedineitherascendingordescendingorderaccordingtotheextentthatasubcategoryreflectsthemagnitudeofvariationinthevariable.Forexample,incomecanbemeasuredeither quantitatively (in dollars and cents) or qualitatively, using subcategories: ‘above average’,‘average’ and ‘below average’. (These categories can also be developed on the basis of quantitativemeasures,forexamplebelow$10000=belowaverage,$10000–$25000=averageandabove$25000=aboveaverage.)Thesubcategory‘aboveaverage’indicatesthatpeoplesogroupedhavemoreincomethanpeopleinthe‘average’category,andpeopleinthe‘average’categoryhavemoreincomethanthoseinthe‘belowaverage’category.Thesesubcategoriesofincomearerelatedtooneanotherintermsofthemagnitudeofpeople’sincome,butthemagnitudeitselfisnotquantifiable,andhencethedifferencebetween‘aboveaverage’and‘average’orbetween‘average’and‘belowaverage’sub-categoriescannotbeascertained.Thesameistrueforothervariablessuchassocioeconomicstatusandattitudesmeasuredonanordinalscale.Therefore,anordinalscalehasalltheproperties/characteristicsofanominalscale,inadditiontoits
own. Subcategories are arranged in order of the magnitude of the property/characteristic. Also, the‘distance’betweenthesubcategoriesisnotequalasthereisnoquantitativeunitofmeasurement.
Theintervalscale
An interval scale has all the characteristics of an ordinal scale; that is, individuals or responsesbelonging to a subcategory have a common characteristic and the subcategories are arranged in anascendingordescendingorder.Inaddition,anintervalscaleusesaunitofmeasurementthatenablestheindividualsorresponsestobeplacedatequallyspacedintervalsinrelationtothespreadofthevariable.Thisscalehasastartingandaterminatingpointandisdividedintoequallyspacedunits/intervals.Thestartingand terminatingpointsand thenumberofunits/intervalsbetween themarearbitraryandvaryfromscaletoscale.Celsius andFahrenheit scales are examples of an interval scale. In theCelsius system the starting
point (considered as the freezing point) is 0°C and the terminating point (considered as the boilingpoint) is 100°C.Thegapbetween the freezing andboilingpoints is divided into100 equally spacedintervals,knownasdegrees.IntheFahrenheitsystemthefreezingpointis32°Fandtheboilingpointis212°F,andthegapbetweenthetwopointsisdividedinto180equallyspacedintervals.Eachdegreeorinterval is ameasurementof temperature– thehigher thedegree, thehigher the temperature.As thestartingandterminatingpointsarearbitrary,theyarenotabsolute;thatis,youcannotsaythat60°Cistwiceashotas30°Cor30°F is three timeshotter than10°F.Thismeans thatwhilenomathematicaloperationcanbeperformedonthereadings,itcanbeperformedonthedifferencesbetweenreadings.Forexample,ifthedifferenceintemperaturebetweentwoobjects,AandB,is15°Candthedifference
in temperature between two other objects, C and D, is 45°C, you can say that the difference intemperaturebetweenCandDisthreetimesgreaterthanthatbetweenAandB.Anattitudetowardsanissue measured on the Thurstone scale is similar. However, the Likert scale does not measure theabsoluteintensityoftheattitudebutsimplymeasuresitinrelationtoanotherperson.Theintervalscaleisrelative;thatis,itplotsthepositionofindividualsorresponsesinrelationtoone
anotherwithrespecttothemagnitudeofthemeasurementvariable.Hence,anintervalscalehasalltheproperties of an ordinal scale, and it has a unit of measurement with an arbitrary starting andterminatingpoint.
Theratioscale
Aratioscalehasallthepropertiesofnominal,ordinalandintervalscalesanditalsohasastartingpointfixedatzero.Therefore,itisanabsolutescale–thedifferencebetweentheintervalsisalwaysmeasuredfrom a zero point. This means the ratio scale can be used for mathematical operations. Themeasurementofincome,age,heightandweightareexamplesofthisscale.Apersonwhois40yearsofageistwiceasoldasa20-year-old.Apersonearning$60000peryearearnsthreetimesthesalaryofapersonearning$20000.
SummaryTheunderstandingandinterpretationofaconceptoraperceptionmayvaryfromrespondenttorespondent,henceitsmeasurementmay not be consistent. A variable has some basis of classification and hence there is far less inconsistency in its meaning andunderstanding.Concepts aremental perceptionswhereas variables aremeasurable either subjectivelyor objectivelyononeof themeasurementscales.Whenyouconvertaconceptintoavariableyouclassifyitonthebasisofmeasurementintocategories,therebyminimisingtheinherentvariabilityinunderstanding.Whenyouareunabletomeasureaconceptdirectly,youneedfirsttoconvertitintoindicatorsandthenintovariables.Theway the required information iscollected inquantitativeandqualitative research is themost significantdifferencebetween
them.Qualitativeresearchmostlyusesdescriptiveornarrativestatementsasthe‘unitsofmeasurement’whereasquantitativeresearchplacesgreateremphasisofmeasuringresponsesononeofthefourmeasurementscales.Thoughqualitativeresearchplacesemphasisondescriptivestatementsindatacollection,atthetimeofanalysis,thesestatementsareclassifiedintocategoriesonthebasisofthemainthemestheycommunicate.Knowledgeofthedifferenttypesofvariablesandthewaytheyaremeasuredplaysacrucialroleinquantitativeresearch.Variables
areimportantinbringingclarityandspecificitytotheconceptualisationofaresearchproblem,totheformulationofhypothesesandtothedevelopmentofaresearchinstrument.Theyaffecthowthedatacanbeanalysed,whatstatisticaltestscanbeappliedtothedata,whatinterpretationscanbemade,howthedatacanbepresentedandwhatconclusionscanbedrawn.Thewayyouaskaquestiondeterminesitscategorisationonameasurementscale,whichinturnaffectshowthedatacanbeanalysed,whatstatisticaltestscanbeappliedtothedata,whatinterpretationscanbemade,howthedatacanbepresentedandwhatconclusionscanbedrawn.Also,thewayavariableismeasuredatthedatacollectionstagetoagreatextentdetermineswhetherastudyisconsideredtobepredominantly‘qualitative’or‘quantitative’innature.Itisimportantforabeginnertounderstandthedifferentwaysinwhichavariablecanbemeasuredandtheimplicationsofthisfor
thestudy.Avariablecanbeclassifiedfromthreeperspectivesthatarenotmutuallyexclusive:causalrelationship,designofthestudyand unit ofmeasurement. From the perspective of causality a variable can be classified into one of four categories: independent,dependent,extraneousandintervening.Fromtheviewpointofstudydesign,therearetwocategoriesofvariable:activeandattribute.Ifwe examine a variable from theperspectiveof theunit ofmeasurement, it canbe classified into categorical and continuousorqualitativeandquantitative.There are four measurement scales used in the social sciences: nominal, ordinal, interval and ratio. Any concept that can be
measuredonthesescalesiscalledavariable.Measurementscalesenablehighlysubjectiveresponses,aswellasresponsesthatcanbemeasuredwithextremeprecision,tobecategorised.Thechoiceofmeasuringavariableonameasurementscaleisdependentuponthepurposeofyourstudyandthewayyouwanttocommunicatethefindingstoreaders.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Imaginethatyouhavebeenaskedtoevaluateyourlecturer.Determinewhichaspectsofteachingyouwouldconsiderimportantanddevelopasetofindicatorsthatmightreflectthese.Self-esteemisadifficultconcepttooperationalise.Thinkabouthowyoumightgoaboutdevelopingasetofindicatorstodeterminevarianceinthelevelofself-esteeminagroupofindividuals.Criticallyexaminethetypologyofvariablesdevelopedinthischapter.Whatchangeswouldyouliketopropose?
CHAPTER6ConstructingHypotheses
Inthischapteryouwilllearnabout:
ThedefinitionofahypothesisThefunctionsofahypothesisinyourresearchHowhypothesesaretestedHowtoformulateahypothesisDifferenttypesofhypothesesandtheirapplicationsHowerrorsinthetestingofahypothesiscanoccurTheuseofhypothesesinqualitativeresearch
Keywords:alternatehypotheses,hunch,hypothesis,hypothesisofpoint-prevalence,null hypothesis, operationalisable, research hypothesis, Type I error, Type II error,unidimensional,valid.
Almosteverygreatstep[inthehistoryofscience]hasbeenmadebythe‘anticipationofnature’,thatis,bytheinventionofhypotheseswhich,thoughverifiable,oftenhadverylittlefoundationtostartwith.(T.H.HuxleycitedinCohen&Nagel1966:197)
Thedefinitionofahypothesis
Thesecondimportantconsiderationintheformulationofaresearchprobleminquantitativeresearchistheconstructionofahypothesis.Hypothesesbringclarity,specificityandfocustoaresearchproblem,but arenot essential for a study.You can conduct a valid investigationwithout constructing a singleformalhypothesis.Ontheotherhand,withinthecontextofaresearchstudy,youcanconstructasmanyhypotheses as you consider to be appropriate. Somebelieve that onemust formulate a hypothesis toundertakeaninvestigation;however,theauthordoesnotholdthisopinion.Hypothesesprimarilyarisefrom a set of ‘hunches’ that are tested through a study and one can conduct a perfectly valid studywithouthavingthesehunchesorspeculations.However,inepidemiologicalstudies,tonarrowthefieldofinvestigation,itisimportanttoformulatehypotheses.The importance of hypotheses lies in their ability to bring direction, specificity and focus to a
researchstudy.Theytellaresearcherwhatspecificinformationtocollect,andtherebyprovidegreaterfocus.Letusimagineyouareattheracesandyouplaceabet.Youbetonahunchthataparticularhorse
willwin.Youwillonlyknowifyourhunchwasrightaftertherace.Takeanotherexample.Supposeyouhaveahunchthattherearemoresmokersthannon-smokersinyourclass.Totestyourhunch,youaskeitherallorjustsomeoftheclassiftheyaresmokers.Youcanthenconcludewhetheryourhunchwasrightorwrong.Now letus takea slightlydifferent example.Supposeyouwork in theareaofpublichealth.Your
clinicalimpressionisthatahigherrateofaparticularconditionprevailsamongpeoplecomingfromaspecificpopulationsubgroup.Youwanttofindouttheprobablecauseofthiscondition.Therecouldbemanycauses.Toexploreeveryconceivablepossibilitywouldrequireanenormousamountoftimeandresources.Hence,tonarrowthechoice,basedonyourknowledgeofthefield,youcouldidentifywhatyouassume tobe themostprobable cause.Youcould thendesign a study to collect the informationneededtoverifyyourhunch.Ifonverificationyouwereabletoconcludethattheassumedcausewastherealcauseofthecondition,yourassumptionwouldhavebeenright.Intheseexamples,youstartedwithasuperficialhunchorassumption.Inonecase(horseracing)you
waited for the event to take place and in the other two instances you designed a study to assess thevalidityofyourassumption,andonlyaftercarefulinvestigationdidyouarriveataconclusionaboutthevalidityofyourassumptions.Hypothesesarebaseduponsimilar logic.Asaresearcheryoudonotknowaboutaphenomenon,a
situation,theprevalenceofaconditioninapopulationorabouttheoutcomeofaprogramme,butyoudohaveahunchtoformthebasisofcertainassumptionsorguesses.Youtestthese,mostlyonebyone,by collecting information thatwill enable you to conclude if your hunchwas right. The verificationprocesscanhaveoneofthreeoutcomes.Yourhunchmayprovetobe:right,partiallyrightorwrong.Without this process of verification, you cannot conclude anything about the validity of yourassumption.Hence, a hypothesis is a hunch, assumption, suspicion, assertion or an idea about a phenomenon,
relationship or situation, the reality or truth of which you do not know. A researcher calls theseassumptions,assertions,statementsorhuncheshypothesesandtheybecomethebasisofanenquiry.Inmoststudiesthehypothesiswillbebaseduponeitherpreviousstudiesoryourownorsomeoneelse’sobservations.Therearemanydefinitionsofahypothesis.According toKerlinger, ‘Ahypothesis isaconjectural
statement of the relationship between two or more variables’ (1986: 17). Webster’s Third NewInternationalDictionary(1976)definesahypothesisas:
aproposition,condition,orprinciplewhich isassumed,perhapswithoutbelief, inorder todrawout its logicalconsequencesandbythismethodto test itsaccordwithfactswhichareknownormaybedetermined.
BlackandChampiondefineahypothesisas ‘a tentativestatementabout something, thevalidityofwhichisusuallyunknown’(1976:126).Inanotherdefinition,Baileydefinesahypothesisas:
apropositionthatisstatedinatestableformandthatpredictsaparticularrelationshipbetweentwo(ormore) variables. In other words, if we think that a relationship exists, we first state it as ahypothesisandthentestthehypothesisinthefield.(1978:35)
AccordingtoGrinnell:
Ahypothesisiswritteninsuchawaythatitcanbeprovenordisprovenbyvalidandreliabledata–itisinordertoobtainthesedatathatweperformourstudy.(1988:200)
Fromtheabovedefinitionsitisapparentthatahypothesishascertaincharacteristics:
1. Itisatentativeproposition.2. Itsvalidityisunknown.3. Inmostcases,itspecifiesarelationshipbetweentwoormorevariables.
Thefunctionsofahypothesis
Whilesomeresearchersbelievethattoconductastudyrequiresahypothesis,havingahypothesisisnotessentialasalreadymentioned.However,ahypothesis is important in termsofbringingclarity to theresearchproblem.Specifically,ahypothesisservesthefollowingfunctions:
Theformulationofahypothesisprovidesastudywithfocus.Ittellsyouwhatspecificaspectsofaresearchproblemtoinvestigate.Ahypothesistellsyouwhatdatatocollectandwhatnottocollect,therebyprovidingfocustothestudy.Asitprovidesafocus,theconstructionofahypothesisenhancesobjectivityinastudy.Ahypothesismayenableyoutoaddtotheformulationoftheory.Itenablesyoutoconcludespecificallywhatistrueorwhatisfalse.
Thetestingofahypothesis
Totestahypothesisyouneedtogothroughaprocess thatcomprises threephases: (1)constructingahypothesis;(2)gatheringappropriateevidence;and(3)analysingevidencetodrawconclusionsastoitsvalidity.Figure6.1showsthisprocessdiagrammatically.Itisonlyafteranalysingtheevidencethatyoucanconcludewhetheryourhunchorhypothesiswastrueorfalse.Whenconcludingaboutahypothesis,conventionally,youspecificallymakeastatementaboutthecorrectnessorotherwiseofahypothesisinthe form of ‘the hypothesis is true’ or ‘the hypothesis is false’. It is therefore imperative that youformulateyourhypothesesclearly,preciselyand inaformthat is testable. Inarrivingataconclusionaboutthevalidityofyourhypothesis,thewayyoucollectyourevidenceisofcentralimportanceanditis therefore essential that your study design, sample, data collection method(s), data analysis andconclusions,andcommunicationoftheconclusionsbevalid,appropriateandfreefromanybias.
FIGURE6.1Theprocessoftestingahypothesis
Thecharacteristicsofahypothesis
There are a number of considerations to keep in mind when constructing a hypothesis, as they areimportantforvalidverification.Thewordingofahypothesisthereforemusthavecertainattributesthatmakeiteasierforyoutoascertainitsvalidity.Theseattributesare:
•Ahypothesisshouldbesimple,specificandconceptuallyclear.Thereisnoplaceforambiguityinthe construction of a hypothesis, as ambiguity will make the verification of your hypothesis almostimpossible. It shouldbe ‘unidimensional’ – that is, it should test onlyone relationshiporhunchat atime.Tobeabletodevelopagoodhypothesisyoumustbefamiliarwiththesubjectarea(theliteraturereviewisof immensehelp).Themore insightyouhave intoaproblem, theeasier it is toconstructahypothesis.Forexample:
Theaverageageofthemalestudentsinthisclassishigherthanthatofthefemalestudents.
The above hypothesis is clear, specific and easy to test. It tells you what you are attempting tocompare(averageageofthisclass),whichpopulationgroupsarebeingcompared(femaleandmalestudents),andwhatyouwanttoestablish(higheraverageageofthemalestudents).Letustakeanotherexample:
Suicideratesvaryinverselywithsocialcohesion.(Black&Champion1976:126)
Thishypothesisisclearandspecific,butalotmoredifficulttotest.Therearethreeaspectsofthishypothesis: ‘suicide rates’; ‘vary inversely’,which stipulates thedirectionof the relationship; and‘socialcohesion’.Tofindoutthesuicideratesandtoestablishwhethertherelationshipisinverseorotherwise are comparatively easy, but to ascertain social cohesion is a lot more difficult. Whatdeterminessocialcohesion?Howcanitbemeasured?Thisproblemmakesitmoredifficult to testthishypothesis.
• Ahypothesisshouldbecapableofverification.Methodsandtechniquesmustbeavailablefordatacollection and analysis. There is no point in formulating a hypothesis if it cannot be subjected toverificationbecausetherearenotechniquestoverifyit.However,thisdoesnotnecessarilymeanthatyoushouldnotformulateahypothesisforwhichtherearenomethodsofverification.Youmight,intheprocessofdoingyourresearch,developnewtechniquestoverifyit.
• A hypothesis should be related to the existing body of knowledge. It is important that yourhypothesisemergesfromtheexistingbodyofknowledge,andthatitaddstoit,asthisisanimportantfunctionofresearch.Thiscanonlybeachievedif thehypothesishas its roots in theexistingbodyofknowledge.
• Ahypothesisshouldbeoperationalisable.Thismeansthat itcanbeexpressedintermsthatcanbemeasured.Ifitcannotbemeasured,itcannotbetestedand,hence,noconclusionscanbedrawn.
Typesofhypothesis
Theoreticallythereshouldbeonlyonetypeofhypothesis,thatistheresearchhypothesis–thebasisofyour investigation. However, because of the conventions in scientific enquiries and because of thewording used in the construction of a hypothesis, hypotheses can be classified into several types.
Broadly,therearetwocategoriesofhypothesis:
1. researchhypotheses;2. alternatehypotheses.
Theformulationofanalternatehypothesisisaconventioninscientificcircles.Itsmainfunctionisto explicitly specify the relationship that will be considered as true in case the research hypothesisproves to be wrong. In a way, an alternate hypothesis is the opposite of the research hypothesis.Conventionally, a null hypothesis, or hypothesis of no difference, is formulated as an alternatehypothesis.Letustakeanexample.Supposeyouwanttotesttheeffectthatdifferentcombinationsofmaternal
andchildhealthservices(MCH)andnutritionalsupplements(NS)haveontheinfantmortalityrate.Totestthis,atwo-by-twofactorialexperimentaldesignisadopted(seeFigure6.2).Thereareseveralwaysofformulatingahypothesis.Forexample:
1. Therewillbenodifferenceinthelevelofinfantmortalityamongthedifferenttreatmentmodalities.
2. TheMCHandNStreatmentgroupswillregisteragreaterdeclineininfantmortalitythantheonlyMCHtreatmentgroup,theonlyNStreatmentgrouporthecontrolgroup.
3. InfantmortalityintheMCHtreatmentgroupwillreachalevelof30/1000overfiveyears.4. DeclineintheinfantmortalityratewillbethreetimesgreaterintheMCHtreatmentgroupthanin
theNSgrouponlyoverfiveyears.
FIGURE6.2Two-by-twofactorialexperimenttostudytherelationshipbetweenMCH,NSandinfantmortality
Let us take another example. Suppose youwant to study the smoking pattern in a community inrelationtogenderdifferentials.Thefollowinghypothesescouldbeconstructed:
1. Thereisnosignificantdifferenceintheproportionofmaleandfemalesmokersinthestudypopulation.
2. Agreaterproportionoffemalesthanmalesaresmokersinthestudypopulation.3. Atotalof60percentoffemalesand30percentofmalesinthestudypopulationaresmokers.4. Therearetwiceasmanyfemalesmokersasmalesmokersinthestudypopulation.
Inbothsetsofexamples,thewaythefirsthypothesishasbeenformulatedindicatesthatthereisnodifferenceeitherintheextentoftheimpactofdifferenttreatmentmodalitiesontheinfantmortalityrateorintheproportionofmaleandfemalesmokers.Whenyouconstructahypothesisstipulatingthatthereis no difference between two situations, groups, outcomes, or the prevalence of a condition orphenomenon,thisiscalledanullhypothesisandisusuallywrittenasH0.Thesecondhypothesis ineachexampleimplies that there isadifferenceeither in theextentof the
impact of different treatmentmodalities on infantmortality or in the proportion ofmale and femalesmokers among the population, though the extent of the difference is not specified.A hypothesis inwhicharesearcherstipulatesthattherewillbeadifferencebutdoesnotspecifyitsmagnitudeiscalledahypothesisofdifference.
FIGURE6.3Typesofhypothesis
A researchermay have enough knowledge about the smoking behaviour of the community or thetreatmentprogrammeanditslikelyoutcomestospeculatealmosttheexactprevalenceofthesituationortheoutcomeofatreatmentprogrammeinquantitativeunits.Examinethethirdhypothesisinbothsetsofexamples:thelevelofinfantmortalityis30/1000andtheproportionoffemaleandmalesmokersis60and30percentrespectively.Thistypeofhypothesisisknownasahypothesisofpoint-prevalence.The fourth hypothesis in both sets of examples speculates a relationship between the impact of
differentcombinationsofMCHandNSprogrammesonthedependentvariable(infantmortality)ortherelationshipbetweentheprevalenceofaphenomenon(smoking)amongdifferentpopulations(maleandfemale). This type of hypothesis stipulates the extent of the relationship in terms of the effect ofdifferenttreatmentgroupsonthedependentvariable(‘threetimesgreaterintheMCHtreatmentgroupthanintheNSgrouponlyoverfiveyears’)ortheprevalenceofaphenomenonindifferentpopulationgroups (‘twice asmany female asmale smokers’).This typeof hypothesis is called ahypothesis ofassociation.NotethatinFigure6.3 thenullhypothesis isalsoclassifiedasahypothesisofnodifferenceunder
‘Researchhypothesis’.Anytypeofhypothesis,includinganullhypothesis,canbecomethebasisofanenquiry. When a null hypothesis becomes the basis of an investigation, it becomes a researchhypothesis.
Errorsintestingahypothesis
Asalreadymentioned,ahypothesisisanassumptionthatmayprovetobeeithercorrectorincorrect.Itis possible to arrive at an incorrect conclusion about a hypothesis for a variety of reasons. Incorrectconclusionsaboutthevalidityofahypothesismaybedrawnif:
thestudydesignselectedisfaulty;thesamplingprocedureadoptedisfaulty;themethodofdatacollectionisinaccurate;theanalysisiswrong;thestatisticalproceduresappliedareinappropriate;ortheconclusionsdrawnareincorrect.
FIGURE6.4TypeIandTypeIIerrorsintestingahypothesis
Any, some or all of these aspects of the research process could be responsible for the inadvertentintroductionoferrorinyourstudy,makingconclusionsmisleading.Hence,inthetestingofahypothesisthereisalwaysthepossibilityoferrorsattributabletothereasonsidentifiedabove.Figure6.4showsthetypesoferrorthatcanresultinthetestingofahypothesis.Hence,indrawingconclusionsaboutahypothesis,twotypesoferrorcanoccur:
Rejectionofanullhypothesiswhenitistrue.ThisisknownasaTypeIerror.Acceptanceofanullhypothesiswhenitisfalse.ThisisknownasaTypeIIerror.
Hypothesesinqualitativeresearch
Oneofthedifferencesinqualitativeandquantitativeresearchisaroundtheimportanceattachedtoandtheextentofuseofhypotheseswhenundertakingastudy.Asqualitativestudiesarecharacterisedbyanemphasis on describing, understanding and exploring phenomena using categorical and subjectivemeasurementprocedures,constructionofhypothesesisneitheradvocatednorpractised.Inaddition,asthedegreeofspecificityneededtotestahypothesisisdeliberatelynotadheredtoinqualitativeresearch,the testing of a hypothesis becomes difficult andmeaningless. This does not mean that you cannotconstructhypothesesinqualitativeresearch;thenon-specificityoftheproblemaswellasmethodsand
procedures make the convention of hypotheses formulation far less practicable and advisable. Evenwithinquantitativestudiestheimportanceattachedtoandthepracticeofformulatinghypothesesvarymarkedly from one academic discipline to another. Fro example, hypotheses are most prevalent inepidemiological research and research relating to the establishment of causality of a phenomenon,where it becomes important to narrow the list of probable causes so that a specific cause-and-effectrelationshipcanbestudied.Inthesocialsciencesformulationofhypothesesismostlydependentontheresearcher and the academic discipline, whereas within an academic discipline it varies markedlybetweenthequantitativeandqualitativeresearchparadigms.
SummaryHypotheses,thoughimportant,arenotessentialforastudy.Aperfectlyvalidstudycanbeconductedwithoutconstructingasinglehypothesis.Hypothesesareimportantforbringingclarity,specificityandfocustoaresearchstudy.Ahypothesisisaspeculativestatementthatissubjectedtoverificationthrougharesearchstudy.Informulatingahypothesisitis
importanttoensurethatitissimple,specificandconceptuallyclear;abletobeverified;rootedinanexistingbodyofknowledge;andabletobeoperationalised.Therearetwobroadtypesofhypothesis:aresearchhypothesisandanalternatehypothesis.Aresearchhypothesiscanbefurther
classified,baseduponthewayitisformulated,asanullhypothesis,ahypothesisofdifference,ahypothesisofpoint-prevalenceandahypothesisofassociation.One of the main differences in qualitative and quantitative research is the extent to which
hypotheses are used and the importance attached to them. In qualitative research, because of thepurposeofan investigationandmethodsused toobtain information,hypothesesarenotusedandalmost no importance is given to them. However, in quantitative research, their use is far moreprevalentthoughitvariesmarkedlyfromoneacademicdisciplinetoanotherandfromresearchertoresearcher.Onthewholeitcanbesaidthat if theaimofastudyis toexplorewhereverylittle isknown,hypothesesareusuallynotformulated;however,ifastudyaimstotestanassertionbywayof causality or association, validate the prevalence of something or establish its existence,hypothesescanbeconstructed.Thetestingofahypothesisbecomesmeaninglessifanyoneoftheaspectsofyourstudy–design,
sampling procedure,method of data collection, analysis of data, statistical procedures applied orconclusions drawn – is faulty or inappropriate. This can result in erroneous verification of ahypothesis:Type Ierroroccurswhereyou rejectanullhypothesiswhen it is trueandshouldnothavebeenrejected;andTypeIIerror is introducedwhereyouacceptanullhypothesiswhen it isfalseandshouldnothavebeenaccepted.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Towhatextentdoyouthinkthattheuseofhypothesesisrelevanttosocialresearch?Formulatetwoorthreehypothesesthatrelatetoyourownareasofinterestandconsiderthefactorsthatmightaffecttheirvalidity.
STEPIIConceptualisingaResearchDesign
Thisoperationalstepincludestwochapters:
Chapter7:TheresearchdesignChapter8:Selectingastudydesign
CHAPTER7TheResearchDesign
Inthischapteryouwilllearnabout:
WhatresearchdesignmeansTheimportantfunctionsofresearchdesignIssuestoconsiderwhendesigningyourownresearchThetheoryofcausalityandtheresearchdesign
Keywords: chance variables, control group, experimental group, extraneousvariables, independent variable, matching, ‘maxmincon’ principle, random error,randomisation,researchdesign,studydesign,treatmentgroup.
Ifyouareclearaboutyourresearchproblem,yourachievementisworthpraising.Youhavecrossedoneofthemostimportantanddifficultsectionsofyourresearchjourney.Havingdecidedwhatyouwanttostudy, younowneed to determinehow you are going to conduct your study.There are a number ofquestionsthatneedtobeansweredbeforeyoucanproceedwithyourjourney.Whatprocedureswillyouadopttoobtainanswerstoresearchquestions?Howwillyoucarryoutthetasksneededtocompletethedifferentcomponentsoftheresearchprocess?Whatshouldyoudoandwhatshouldyounotdointheprocessofundertakingthestudy?Basically,answerstothesequestionsconstitutethecoreofaresearchdesign.
Whatisaresearchdesign?
A research design is a plan, structure and strategy of investigation so conceived as to obtainanswerstoresearchquestionsorproblems.Theplanisthecompleteschemeorprogrammeoftheresearch. It includesanoutlineofwhat the investigatorwilldo fromwriting thehypothesesandtheiroperationalimplicationstothefinalanalysisofdata.(Kerlinger1986:279)
A traditional research design is a blueprint or detailed plan for how a research study is to becompleted—operationalizingvariablessotheycanbemeasured,selectingasampleof interest tostudy,collectingdatatobeusedasabasisfortestinghypotheses,andanalysingtheresults.(Thyer
1993:94)
A research design is a procedural plan that is adopted by the researcher to answer questions validly,objectively,accuratelyandeconomically.AccordingtoSelltiz,DeutschandCook,‘Aresearchdesignisthe arrangement of conditions for collection and analysis of data in amanner that aims to combinerelevance to the researchpurposewitheconomy inprocedure’ (1962:50).Througha researchdesignyou decide for yourself and communicate to others your decisions regardingwhat study design youproposetouse,howyouaregoingtocollectinformationfromyourrespondents,howyouaregoingtoselectyourrespondents,howtheinformationyouaregoingtocollectistobeanalysedandhowyouaregoing to communicateyour findings. In addition,youwill need todetail inyour researchdesign therationale and justification for each decision that shapes your answers to the ‘how’ of the researchjourney. In presenting your rationale and justification you need to support them critically from theliterature reviewed.Youalsoneed toassureyourselfandothers that thepathyouhaveproposedwillyieldvalidandreliableresults.
Thefunctionsofaresearchdesign
The above definitions suggest that a research design has twomain functions.The first relates to theidentification and/or development of procedures and logistical arrangements required to undertake astudy,andthesecondemphasisestheimportanceofqualityintheseprocedurestoensuretheirvalidity,objectivityandaccuracy.Hence,througharesearchdesignyou:
conceptualiseanoperationalplantoundertakethevariousproceduresandtasksrequiredtocompleteyourstudy;ensurethattheseproceduresareadequatetoobtainvalid,objectiveandaccurateanswerstotheresearchquestions.Kerlingercallsthisfunctionthecontrolofvariance(1986:280).
Letustakethefirstofthesefunctions.Theresearchdesignshoulddetailforyou,yoursupervisorandotherreadersalltheproceduresyouplantouseandthetasksyouaregoingtoperformtoobtainanswerstoyourresearchquestions.Oneof themost importantrequirementsofa researchdesign is tospecifyeverythingclearly soa readerwillunderstandwhatprocedures to followandhow to follow them.Aresearchdesign,therefore,shoulddothefollowing:
Namethestudydesignperse–thatis,‘cross-sectional’,‘before-and-after’,‘comparative’,‘controlexperiment’or‘randomcontrol’.Providedetailedinformationaboutthefollowingaspectsofthestudy:
Whowillconstitutethestudypopulation?Howwillthestudypopulationbeidentified?Willasampleorthewholepopulationbeselected?Ifasampleisselected,howwillitbecontacted?Howwillconsentbesought?Whatmethodofdatacollectionwillbeusedandwhy?Inthecaseofaquestionnaire,wherewilltheresponsesbereturned?Howshouldrespondentscontactyouiftheyhavequeries?
Inthecaseofinterviews,wherewilltheybeconducted?Howwillethicalissuesbetakencareof?
Chapter8describessomeofthecommonlyusedstudydesigns.Therestofthetopicsthatconstitutearesearchdesignarecoveredinthesubsequentchapters.
Thetheoryofcausalityandtheresearchdesign
Nowlet’sturntothesecondfunctionoftheresearchdesign–ensuringthattheproceduresundertakenareadequatetoobtainvalid,objectiveandaccurateanswerstotheresearchquestions.Toensurethis,itisimportantthatyouselectastudydesignthathelpsyoutoisolate,eliminateorquantifytheeffectsofdifferentsetsofvariable influencing the independentvariable.Tohelpexplain this,we lookata fewexamples.Suppose youwant to find out the effectiveness of amarriage counselling service provided by an
agency– that is, the extent towhich the servicehasbeenable to resolve themaritalproblemsof itsclients. In studying such relationships you must understand that in real life there are many outsidefactorsthatcaninfluencetheoutcomeofyourintervention.Forexample,duringvisitstoyouragencyforcounselling,yourclientmaygetabetterjob.Ifsomeofthemaritalproblemscameaboutbecauseofeconomic hardship, and if the problem ofmoney is now solved, itmay be a factor in reducing themaritalproblems.Ontheotherhand,ifaclientloseshis/herjob,theincreaseintheeconomicproblemsmayeitherintensifyorlessenthemaritalproblems;thatis,forsomecouplesaperceivedfinancialthreatmayincreasemaritalproblems,whereas,forothers,itmaycreatemoreclosenessbetweenpartners.Insomesituations,animprovementinamarriagemayhaveverylittletodowiththecounsellingreceived,comingaboutalmostentirelybecauseofachangeineconomiccircumstances.Othereventssuchasthebirthofachild toacoupleoracouple’s independent‘self-realisation’, independentlyarrivedat,mayalsoaffect theextentandnatureofmaritalproblems.Figure7.1 listsotherpossible factorsunder thecategoryofextraneousvariables.Thisdoesnotexhaustthelistbyanymeans.
FIGURE7.1FactorsaffectingtherelationshipbetweenacounsellingserviceandtheextentofmaritalproblemsContinuing the example of marriage and counselling, there are sets of factors that can affect the
relationshipbetweencounsellingandmarriageproblems,andeachisadefinedcategoryofvariables:
1. Counsellingperse.2. Allthefactorsotherthancounsellingthataffectthemaritalproblems.3. Theoutcome–thatis,thechangeorotherwiseintheextentofthemaritalproblems.4. Sometimes,thevariationinresponsetoquestionsaboutmaritalproblemscanbeaccountedforby
themoodofrespondentsorambiguityinthequestions.Somerespondentsmayeitheroverestimateorunderestimatetheirmaritalproblemsbecauseoftheirstateofmindatthetime.Orsomerespondents,inspiteofbeinginexactlythesamesituation,mayrespondtonon-specificorambiguousquestionsdifferently,accordingtohowtheyinterpretthequestion.
As already explained inChapter5, any variable that is responsible for bringing about a change iscalledanindependentvariable.Inthisexample,thecounsellingisanindependentvariable.Whenyoustudyacause-and-effect relationship,usuallyyoustudy the impactofonlyone independentvariable.Occasionallyyoumaystudytheimpactoftwoindependentvariables,or(veryrarely)morethantwo,butthesestudydesignsaremorecomplex.For this example counsellingwas the assumed cause of change in the extent ofmarital problems;
hence,theextentofmaritalproblemsisthedependentvariable,asthechangeinthedegreeofmaritalproblemswasdependentuponcounselling.All other factors that affect the relationship between marital problems and counselling are called
extraneousvariables.Inthesocialsciences,extraneousvariablesoperateineverystudyandcannotbeeliminated.However,theycanbecontrolledtosomeextent.(Someofthemethodsforcontrollingthemare described later in this chapter.)Nevertheless, it is possible to find out the impact attributable toextraneousvariables.Thisisdonewiththeintroductionofacontrolgroupinthestudydesign.Thesolefunction of a control group is to quantify the impact of extraneous variables on the dependentvariable(s).Changes in the dependent variable, because of the respondent’s state ofmoodor ambiguity in the
research instrument, are called randomvariables orchancevariables. The error thus introduced iscalledthechanceorrandomerror.Inmostcasestheneteffectofchancevariablesisconsideredtobenegligibleasrespondentswhooverreporttendtocanceloutthosewhounderreport.Thesameappliestoresponsestoambiguousquestionsinaresearchinstrument.Henceinanycausalrelationship,changesinthedependentvariablemaybeattributedtothreetypes
ofvariable:
Letustakeanotherexample.Supposeyouwanttostudytheimpactofdifferentteachingmodelsonthelevelofcomprehensionofstudentsforwhichyouadoptacomparativestudydesign.Inthisstudy,the change in the level of comprehension, in addition to the teachingmodels, can be attributed to anumberofotherfactors,someofwhichareshowninFigure7.2:
[changeinlevelofcomprehension]=
[changeattributabletotheteachingmodel]±[changeattributabletoextraneousvariables]±[changeattributabletochancevariables]
Infact,inanystudythatattemptstoestablishacausalrelationship,youwilldiscoverthattherearethree sets of variables operating to bring about a change in the dependent variable. This can be
expressedasanequation:
[changeintheoutcomevariable]=
[changebecauseofthechancevariable]±[changebecauseofextraneousvariables]±[changebecauseofchanceorrandomvariables]
orinotherwords:
[changeinthedependentvariable]=
[changeattributabletotheindependentvariable]±[changeattributabletoextraneousvariables]±[changeattributabletochancevariables]
FIGURE7.2Therelationshipbetweenteachingmodelsandcomprehension
orintechnicalterms:
[totalvariance]=
[varianceattributabletotheindependentvariable]±[varianceattributabletoextraneousvariables]±[randomorchancevariance]
Itcanalsobeexpressedgraphically(Figure7.3).Asthetotalchangemeasuresthecombinedeffectofallthreecomponentsitisdifficulttoisolatethe
individualimpactofeachofthem(seeFigure7.3).Sinceyouraimasaresearcheristodeterminethechangethatcanbeattributedtotheindependentvariable,youneedtodesignyourstudytoensurethattheindependentvariablehasthemaximumopportunitytohaveitsfulleffectonthedependentvariable,while theeffects thatareattributed toextraneousandchancevariablesareminimised (ifpossible)orquantified or eliminated. This is what Kerlinger (1986: 286) calls the ‘maxmincon’ principle ofvariance.Oneofthemostimportantquestionsis:howdoweminimisetheeffectattributabletoextraneousand
chancevariables?Theansweristhatinmostsituationswecannot;however, itcanbequantified.The
solepurposeofhavingacontrolgroup,asmentionedearlier,istomeasurethechangethatisaresultofextraneous variables. The effect of chance variables is often assumed to be none or negligible. Asdiscussed,chancevariationcomesprimarilyfromtwosources:respondentsandtheresearchinstrument.It is assumed that if some respondents affect the dependent variable positively, others will affect itnegatively.Forexample,ifsomerespondentsareextremelypositiveintheirattitudetowardsanissue,beingveryliberalorpositivelybiased,thereareboundtobeotherswhoareextremelynegative(beingveryconservativeornegativelybiased).Hence, theytendtocanceleachotheroutsotheneteffect isassumed to be zero. However, if in a study population most individuals are either negatively orpositively biased, a systematic error in the findings will be introduced. Similarly, if a researchinstrumentisnotreliable(i.e.itisnotmeasuringcorrectlywhatitissupposedtomeasure),asystematicbiasmaybeintroducedintothestudy.
FIGURE7.3TheproportionattributabletothethreecomponentsmayvarymarkedlyInthephysicalsciencesaresearchercancontrolextraneousvariablesasexperimentsareusuallydone
inalaboratory.Bycontrast,inthesocialsciences,thelaboratoryissociety,overwhichtheresearcherlackscontrol.Sincenoresearcherhascontroloverextraneousvariables, theireffect,asmentioned, inmostsituationscannotbeminimised.Thebestoption is toquantify their impact through theuseofacontrol group, though the introduction of a control group creates the problem of ensuring that theextraneousvariableshaveasimilareffectonbothcontrolandexperimentalgroups.Insomesituationstheirimpactcanbeeliminated(thisispossibleonlywhereoneortwovariableshaveamarkedimpacton the dependent variable). There are twomethods used to ensure that extraneous variables have asimilareffectoncontrolandexperimentalgroupsandtwomethodsforeliminatingextraneousvariables:
FIGURE7.4Buildingintothedesign
1. Ensurethatextraneousvariableshaveasimilarimpactoncontrolandexperimentalgroups.Itisassumedthatiftwogroupsarecomparable,theextenttowhichtheextraneousvariableswill
affectthedependentvariablewillbesimilarinbothgroups.Thefollowingtwomethodsensurethatthecontrolandexperimentalgroupsarecomparablewithoneanother:
(a)Randomisation–Ensuresthatthetwogroupsarecomparablewithrespecttothevariable(s).Itisassumedthatif thegroupsarecomparable, theextenttowhichextraneousvariablesaregoingtoaffectthedependentvariableisthesameineachgroup.
(b)Matching –Anotherway of ensuring that the two groups are comparable so that the effect ofextraneousvariableswillbethesameinbothgroups(discussedinChapter8).
2. Eliminateextraneousvariable(s).Sometimesitispossibletoeliminatetheextraneousvariableortobuilditintothestudydesign.Thisisusuallydonewhenthereisstrongevidencethattheextraneousvariablehasahighcorrelationwiththedependentvariable,orwhenyouwanttoisolatetheimpactoftheextraneousvariable.Therearetwomethodsusedtoachievethis:
(a)Buildtheaffectingvariableintothedesignofthestudy–Toexplainthisconceptletustakeanexample.Supposeyouwanttostudytheimpactofmaternalhealthservicesontheinfantmortalityofapopulation.Itcanbeassumedthatthenutritionalstatusofchildrenalsohasamarkedeffectoninfantmortality.Tostudy the impactofmaternalhealthservicesperse,youadopta two-by-twofactorialdesignasexplainedinFigure7.4.Inthiswayyoucanstudytheimpactoftheextraneousvariablesseparatelyandinteractivelywiththeindependentvariable.
(b)Eliminatethevariable–Tounderstand this, letus takeanotherexample.Supposeyouwant tostudytheimpactofahealtheducationprogrammeontheattitudestowards,andbeliefsabout,thecausation and treatment of a certain illness among non-indigenous Australians and indigenousAustralianslivinginaparticularcommunity.Asattitudesandbeliefsvarymarkedlyfromculturetoculture,studyingnon-indigenousAustraliansandindigenousAustraliansasonegroupwillnotprovideanaccuratepicture.Insuchstudiesit isappropriate toeliminatetheculturalvariationinthe study population by selecting and studying the populations separately or by constructingculture-specificcohortsatthetimeofanalysis.
SummaryInthischapteryouhavelearntaboutthefunctionsofaresearchdesign.Aresearchdesignservestwoimportantfunctions:(1)todetailtheproceduresforundertakingastudy;and(2) toensure that, in thecaseofcausality, the independentvariablehas themaximumopportunitytohaveitseffectonthedependentvariablewhiletheeffectofextraneousandchancevariablesisminimised.Intermsofthefirstfunction,aresearchdesignshouldoutlinethelogisticaldetailsofthewholeprocessoftheresearchjourney.Youneedtospelloutindetailwhattypeofstudydesignperseyouareproposingtouseandwhy,whoaregoingtobeyourrespondentsandhowtheywillbeselected,fromhowmanyyouareproposingtogettheneededinformation,howtheinformationwillbecollectedbyyouandhowyou are going to analyse the information.For each aspect youneed to provide your rationale and justification and as far aspossiblesupportthemfromtheliteraturereviewed.Through the second function, ‘Control of variance’, when establishing association or causality, it ensures your supervisor and
readers that you have set up your study in such away that your independent variable has themaximum chance of affecting thedependent variable and that the effects of extraneous and chance variables are minimised, quantified and/or controlled (the‘maxmincon’principleofvariance).Astudywithoutacontrolgroupmeasuresthetotalchange(changeattributabletoindependentvariable±extraneousvariables±
chancevariables)inaphenomenonorsituation.Thepurposeofintroducingacontrolgroupistoquantifytheimpactofextraneousandchancevariables.Thestudydesignisapartoftheresearchdesign.Itisthedesignofthestudyperse,whereastheresearchdesignalsoincludesother
detailsrelatedtothecarryingoutofthestudy.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Whatarethemainfunctionsofaresearchdesign?Whyisitimportanttohavearesearchdesignbeforeundertakingastudy?Provideexamplesfromyourownareaofstudytoillustratethemainvariablesintermsofcausality(youmayfinditusefultoreferbacktoChapter5).Identifyoneortwoexamplesfromanareathatinterestsyoutodemonstratehowthe‘maxmincon’principleofvariancecanbeapplied.
CHAPTER8SelectingaStudyDesign
Inthischapteryouwilllearnabout:
ThedifferencesbetweenquantitativeandqualitativestudydesignsCommonstudydesignsinquantitativeresearchandwhentousethemCommonstudydesigninqualitativeresearchandwhentousethemThestrengthsandweaknessesofdifferentstudydesigns
Keywords:actionresearch,after-onlydesign,before-and-afterstudydesign,blindstudies, case studies, cohort studies, control studies, cross-sectional study design,double-blind studies, experimental study design, feminist research, focus studies,longitudinal studies, non-experimental studies, panel studies, prospective studydesign, quasi-experimental studies, reflective journal, retrospective studies, semi-experimentalstudies,trendstudies.
Differencesbetweenquantitativeandqualitativestudydesigns
Inthischapterwewilldiscusssomeofthemostcommonlyusedstudydesignsinbothquantitativeandqualitative research. Overall, there are many more study designs in quantitative research than inqualitativeresearch.Quantitativestudydesignsarespecific,wellstructured,havebeentestedfortheirvalidity and reliability, and can be explicitly defined and recognised. Study designs in qualitativeresearcheitherdonothavetheseattributesorhavethemtoalesserdegree.Theyarelessspecificandprecise,anddonothavethesamestructuraldepth.Differencesinphilosophicalperspectivesineachparadigmcombinedwiththeaimsofastudy,toa
largeextent,determinethefocus,approachandmodeofenquirywhich,inturn,determinethestructuralaspects of a studydesign. Themain focus in qualitative research is to understand, explain, explore,discover and clarify situations, feelings, perceptions, attitudes, values, beliefs and experiences of agroupofpeople.Thestudydesignsarethereforeoftenbasedondeductiveratherthaninductivelogic,are flexible and emergent in nature, and are often non-linear and non-sequential in theiroperationalisation.Thestudydesignsmainlyentailtheselectionofpeoplefromwhomtheinformation,throughanopenframeofenquiry,isexploredandgathered.Theparametersofthescopeofastudy,and
informationgatheringmethodsandprocesses,areoftenflexibleandevolving;hence,mostqualitativedesigns are not as structured and sequential as quantitative ones. On the other hand, in quantitativeresearch, themeasurementandclassificationrequirementsof the information that isgathereddemandthatstudydesignsaremorestructured,rigid,fixedandpredeterminedintheirusetoensureaccuracyinmeasurementandclassification.Inqualitativestudiesthedistinctionbetweenstudydesignsandmethodsofdatacollectionisfarless
clear.Quantitativestudydesignshavemoreclarityanddistinctionbetweendesignsandmethodsofdatacollection. In qualitative research there is an overlap between the two. Some designs are basicallymethodsofdatacollection.Forexample,in-depthinterviewingisadesignaswellasamethodofdatacollectionandsoareoralhistoryandparticipantobservation.One of themost distinguishing features of qualitative research is the adherence to the concept of
respondent concordance whereby you as a researcher make every effort to seek agreement of yourrespondents with your interpretation, presentation of the situations, experiences, perceptions andconclusions. In quantitative research respondent concordance does not occupy an important place.Sometimes it is assumed to be achieved by circulating or sharing the findings with those whoparticipatedinthestudy.The‘power-gap’betweentheresearcherandthestudypopulationinqualitativeresearchisfarsmaller
than in quantitative research because of the informality in structure and situation in which data iscollected.In quantitative research enough detail about a study design is provided for it to be replicated for
verificationandreassurance.Inqualitativeresearchlittleattentionispaidtostudydesignsortheotherstructuralaspectsofastudy,hencethereplicationofastudydesignbecomesalmost impossible.Thisleads to the inability of the designs to produce findings that can be replicated. Findings throughquantitativestudydesignscanbe replicatedandretestedwhereas thiscannotbeeasilydonebyusingqualitativestudydesigns.Another difference in the designs in qualitative and quantitative studies is the possibility of
introducing researcher bias. Because of flexibility and lack of control it is more difficult to checkresearcherbiasinqualitativestudies.Study designs in each paradigm are appropriate for finding different things. Study designs in
qualitativeresearcharemoreappropriateforexploringthevariationanddiversityinanyaspectofsociallife,whereasinquantitativeresearchtheyaremoresuitedtofindingouttheextentofthisvariationanddiversity. If your interest is in studying values, beliefs, understandings, perceptions, meanings, etc.,qualitativestudydesignsaremoreappropriateastheyprovideimmenseflexibility.Ontheotherhand,ifyour focus is tomeasure themagnitudeof thatvariation, ‘howmanypeoplehaveaparticularvalue,belief, etc.?’, the quantitative designs are more appropriate. For good quantitative research it isimportant thatyoucombinequantitativeskillswithqualitativeoneswhenascertaining thenatureandextentofdiversityandvariationinaphenomenon.Intheauthor’sopinion,thequalitative–quantitative–qualitativeapproachtoresearchiscomprehensiveandworthconsideration.Thisinvolvesstartingwithqualitative methods to determine the spread of diversity, using quantitative methods to quantify thespreadandthengoingbacktoqualitativetoexplaintheobservedpatterns.Asalreadystated,theauthordoes not recommend your locking yourself into either the qualitative or quantitative paradigm and,though you may have your preference, it is the purpose that should determine the choice betweenquantitative and qualitative study designs. If you already know (from previous studies or practiceknowledge) the nature of diversity in any area of interest to you, knowledge about its extent can bedeterminedonlybyusingquantitativemethods.Inmostcaseswhereyouwanttoexploreboth,youneedtousemethodsthatfallinthedomainofbothparadigms.
Studydesignsinquantitativeresearch
Someofthecommonlyuseddesignsinquantitativestudiescanbeclassifiedbyexaminingthemfromthreedifferentperspectives:
1. thenumberofcontactswiththestudypopulation;2. thereferenceperiodofthestudy;3. thenatureoftheinvestigation.
Every study design can be classified from each one of these perspectives. These perspectives arearbitrary bases of classification; hence, the terminology used to describe them is not universal.However, thenamesof thedesignswithineachclassificationbaseareuniversallyused.Note that thedesignswithineachcategoryaremutuallyexclusive; that is, ifaparticularstudy iscross-sectional innature it cannotbeat the same timeabefore-and-afteror a longitudinalstudy, but it canbeanon-experimental or experimental study, as well as a retrospective study or a prospective study. SeeFigure8.1.Anothersectionhasbeenadded to the threesections listedabove titled ‘Others–somecommonly
usedstudydesigns’.Thissectionincludessomecommonlyuseddesignswhicharebasedonacertainphilosophyormethodology,andwhichhaveacquiredtheirownnames.
Studydesignsbasedonthenumberofcontacts
Basedonthenumberofcontactswiththestudypopulation,designscanbeclassifiedintothreegroups:
1. cross-sectionalstudies;2. before-and-afterstudies;3. longitudinalstudies.
FIGURE8.1Typesofstudydesign
Thecross-sectionalstudydesign
Cross-sectionalstudies,alsoknownasone-shotorstatusstudies,arethemostcommonlyuseddesignin the social sciences. This design is best suited to studies aimed at finding out the prevalence of aphenomenon,situation,problem,attitudeorissue,bytakingacross-sectionofthepopulation.Theyareusefulinobtaininganoverall‘picture’asitstandsatthetimeofthestudy.Theyare‘designedtostudysomephenomenonbytakingacross-sectionofitatonetime’(Babbie1989:89).Suchstudiesarecross-sectionalwithregardtoboththestudypopulationandthetimeofinvestigation.Across-sectionalstudyisextremelysimpleindesign.Youdecidewhatyouwanttofindoutabout,
identifythestudypopulation,selectasample(ifyouneedto)andcontactyourrespondentstofindoutthe required information. For example, a cross-sectional designwould be themost appropriate for astudyofthefollowingtopics:
TheattitudeofthestudypopulationtowardsuraniummininginAustralia.Thesocioeconomic–demographiccharacteristicsofimmigrantsinWesternAustralia.TheincidenceofHIV-positivecasesinAustralia.Thereasonsforhomelessnessamongyoungpeople.
Thequalityassuranceofaserviceprovidedbyanorganisation.Theimpactofunemploymentonstreetcrime(thiscouldalsobeabefore-and-afterstudy).Therelationshipbetweenthehomeenvironmentandtheacademicperformanceofachildatschool.Theattitudeofthecommunitytowardsequityissues.Theextentofunemploymentinacity.Consumersatisfactionwithaproduct.Theeffectivenessofrandombreathtestinginpreventingroadaccidents(thiscouldalsobeabefore-and-afterstudy).Thehealthneedsofacommunity.Theattitudesofstudentstowardsthefacilitiesavailableintheirlibrary.
Asthesestudiesinvolveonlyonecontactwiththestudypopulation,theyarecomparativelycheaptoundertakeandeasytoanalyse.However,theirbiggestdisadvantageisthattheycannotmeasurechange.Tomeasurechangeitisnecessarytohaveatleasttwodatacollectionpoints–thatis,atleasttwocross-sectionalstudies,attwopointsintime,onthesamepopulation.
Thebefore-and-afterstudydesign
Themainadvantageofthebefore-and-afterdesign(alsoknownasthepre-test/post-testdesign)isthatitcanmeasurechange inasituation,phenomenon, issue,problemorattitude. It is themostappropriatedesign formeasuring the impact or effectiveness of a programme.A before-and-after design can bedescribedas two setsof cross-sectionaldata collectionpointson the samepopulation to findout thechange in the phenomenon or variable(s) between two points in time. The change is measured bycomparingthedifferenceinthephenomenonorvariable(s)beforeandaftertheintervention(seeFigure8.2).
FIGURE8.2Before-and-after(pre-test/post-test)studydesignAbefore-and-afterstudyiscarriedoutbyadoptingthesameprocessasacross-sectionalstudyexcept
that it comprises two cross-sectional data sets, the second being undertaken after a certain period.Depending upon how it is set up, a before-and-after study may be either an experiment or a non-experiment.Itisoneofthemostcommonlyuseddesignsinevaluationstudies.Thedifferencebetweenthe two sets of data collection pointswith respect to the dependent variable is considered to be theimpactoftheprogramme.Thefollowingareexamplesoftopicsthatcanbestudiedusingthisdesign:
Theimpactofadministrativerestructuringonthequalityofservicesprovidedbyanorganisation.Theeffectivenessofamarriagecounsellingservice.Theimpactofsexeducationonsexualbehaviouramongschoolchildren.Theeffectofadrugawarenessprogrammeontheknowledgeabout,anduseof,drugsamongyoungpeople.Theimpactofincentivesontheproductivityofemployeesinanorganisation.Theimpactofincreasedfundingonthequalityofteachinginuniversities.Theimpactofmaternalandchildhealthservicesontheinfantmortalityrate.Theeffectofrandombreathtestingonroadaccidents.Theeffectofanadvertisementonthesaleofaproduct.
Themainadvantageofbefore-and-afterdesignisitsabilitytomeasurechangeinaphenomenonortoassess the impact of an intervention. However, there can be disadvantages which may not occur,individuallyorcollectively,ineverystudy.Theprevalenceofaparticulardisadvantage(s)isdependentupon the nature of the investigation, the study population and themethod of data collection. Thesedisadvantagesincludethefollowing:
Astwosetsofdatamustbecollected,involvingtwocontactswiththestudypopulation,thestudyismoreexpensiveandmoredifficulttoimplement.Italsorequiresalongertimetocomplete,particularlyifyouareusinganexperimentaldesign,asyouwillneedtowaituntilyourinterventioniscompletedbeforeyoucollectthesecondsetofdata.Insomecasesthetimelapsebetweenthetwocontactsmayresultinattritioninthestudypopulation.Itispossiblethatsomeofthosewhoparticipatedinthepre-testmaymoveoutoftheareaorwithdrawfromtheexperimentforotherreasons.Oneofthemainlimitationsofthisdesign,initssimplestform,isthatasitmeasurestotalchange,youcannotascertainwhetherindependentorextraneousvariablesareresponsibleforproducingchangeinthedependentvariable.Also,itisnotpossibletoquantifythecontributionofindependentandextraneousvariablesseparately.Ifthestudypopulationisveryyoungandifthereisasignificanttimelapsebetweenthebefore-and-aftersetsofdatacollection,changesinthestudypopulationmaybebecauseitismaturing.Thisisparticularlytruewhenyouarestudyingyoungchildren.Theeffectofthismaturation,ifitissignificantlycorrelatedwiththedependentvariable,isreflectedatthe‘after’observationandisknownasthematurationeffect.Sometimestheinstrumentitselfeducatestherespondents.Thisisknownasthereactiveeffectoftheinstrument.Forexample,supposeyouwanttoascertaintheimpactofaprogrammedesignedtocreateawarenessofdrugsinapopulation.Todothis,youdesignaquestionnairelistingvariousdrugsandaskingrespondentstoindicatewhethertheyhaveheardofthem.Atthepre-teststagearespondent,whileansweringquestionsthatincludethenamesofthevariousdrugs,isbeingmadeawareofthem,andthiswillbereflectedinhis/herresponsesatthepost-teststage.Thus,theresearchinstrumentitselfhaseducatedthestudypopulationand,hence,hasaffectedthedependentvariable.Anotherexampleofthiseffectisastudydesignedtomeasuretheimpactofafamilyplanningeducationprogrammeonrespondents’awarenessofcontraceptivemethods.Moststudiesdesignedtomeasuretheimpactofaprogrammeonparticipants’awarenessfacethedifficultythatachangeinthelevelofawareness,tosomeextent,maybebecauseofthisreactiveeffect.Anotherdisadvantagethatmayoccurwhenyouusearesearchinstrumenttwicetogaugetheattitudeofapopulationtowardsanissueisapossibleshiftinattitudebetweenthetwopointsofdatacollection.Sometimespeoplewhoplacethemselvesattheextremepositionsofa
measurementscaleatthepre-teststagemay,foranumberofreasons,shifttowardsthemeanatthepost-teststage(seeFigure8.3).Theymightfeelthattheyhavebeentoonegativeortoopositiveatthepre-teststage.Therefore,themereexpressionofanattitudeinresponsetoaquestionnaireorinterviewhascausedthemtothinkaboutandaltertheirattitudeatthetimeofthepost-test.Thistypeofeffectisknownastheregressioneffect.
FIGURE8.3Theregressioneffect
Thelongitudinalstudydesign
Thebefore-and-afterstudydesignisappropriateformeasuringtheextentofchangeinaphenomenon,situation, problem, attitude, and so on, but is less helpful for studying the pattern of change. Todeterminethepatternofchangeinrelationtotime,alongitudinaldesignisused;forexample,whenyouwish tostudy theproportionofpeopleadoptingaprogrammeoveraperiod.Longitudinalstudiesarealso useful when you need to collect factual information on a continuing basis. You may want toascertainthetrendsinthedemandforlabour,immigration,changesintheincidenceofadiseaseorinthemortality,morbidityandfertilitypatternsofapopulation.Inlongitudinalstudiesthestudypopulationisvisitedanumberoftimesatregularintervals,usually
overalongperiod,tocollecttherequiredinformation(seeFigure8.4).Theseintervalsarenotfixedsotheirlengthmayvaryfromstudytostudy.Intervalsmightbeasshortasaweekorlongerthanayear.Irrespectiveofthesizeoftheinterval,thetypeofinformationgatheredeachtimeisidentical.Althoughthedatacollectedisfromthesamestudypopulation,itmayormaynotbefromthesamerespondents.Alongitudinalstudycanbeseenasaseriesofrepetitivecross-sectionalstudies.
FIGURE8.4ThelongitudinalstudydesignLongitudinal studies have many of the same disadvantages as before-and-after studies, in some
instancestoanevengreaterdegree.Inaddition,longitudinalstudiescansufferfromtheconditioningeffect.Thisdescribesasituationwhere,ifthesamerespondentsarecontactedfrequently,theybegintoknowwhatisexpectedofthemandmayrespondtoquestionswithoutthought,ortheymayloseinterest
intheenquiry,withthesameresult.Themainadvantageofalongitudinalstudyisthatitallowstheresearchertomeasurethepatternof
change and obtain factual information, requiring collection on a regular or continuing basis, thusenhancingitsaccuracy.
Studydesignsbasedonthereferenceperiod
Thereferenceperiod refers to the time-frame inwhichastudy isexploringaphenomenon,situation,eventorproblem.Studiesarecategorisedfromthisperspectiveas:
retrospective;prospective;retrospective–prospective.
Theretrospectivestudydesign
Retrospective studies investigate aphenomenon, situation,problemor issue thathashappened in thepast.Theyareusuallyconductedeitheronthebasisofthedataavailableforthatperiodoronthebasisof respondents’ recallof thesituation(Figure8.5a).Forexample, studiesconductedon the followingtopicsareclassifiedasretrospectivestudies:
ThelivingconditionsofAboriginalandTorresStraitIslanderpeoplesinAustraliaintheearlytwentiethcentury.TheutilisationoflandbeforetheSecondWorldWarinWesternAustralia.AhistoricalanalysisofmigratorymovementsinEasternEuropebetween1915and1945.Therelationshipbetweenlevelsofunemploymentandstreetcrime.
Theprospectivestudydesign
Prospective studies refer to the likely prevalence of a phenomenon, situation, problem, attitude oroutcomeinthefuture(Figure8.5b).Suchstudiesattempttoestablishtheoutcomeofaneventorwhatislikelytohappen.Experimentsareusuallyclassifiedasprospectivestudiesastheresearchermustwaitfor an intervention to register its effect on the study population. The following are classified asprospectivestudies:
Todetermine,underfieldconditions,theimpactofmaternalandchildhealthservicesonthelevelofinfantmortality.Toestablishtheeffectsofacounsellingserviceontheextentofmaritalproblems.Todeterminetheimpactofrandombreathtestingonthepreventionofroadaccidents.Tofindouttheeffectofparentalinvolvementonthelevelofacademicachievementoftheirchildren.TomeasuretheeffectsofachangeinmigrationpolicyontheextentofimmigrationinAustralia.
Theretrospective–prospectivestudydesign
Retrospective–prospectivestudiesfocusonpasttrendsinaphenomenonandstudyitintothefuture.Part of the data is collected retrospectively from the existing records before the intervention isintroducedandthenthestudypopulationisfollowedtoascertaintheimpactoftheintervention(Figure8.5c).
FIGURE8.5(a)Retrospectivestudydesign;(b)prospectivestudydesign;(c)retrospective–prospectivestudydesign.A study is classified under this categorywhen youmeasure the impact of an interventionwithout
havingacontrolgroup.Infact,mostbefore-and-afterstudies,ifcarriedoutwithouthavingacontrol–wherethebaselineisconstructedfromthesamepopulationbeforeintroducingtheintervention–willbeclassified as retrospective–prospective studies.Trend studies,which become the basis of projections,fallintothiscategorytoo.Someexamplesofretrospective–prospectivestudiesare:
Theeffectofrandombreathtestingonroadaccidents.Theimpactofincentivesontheproductivityoftheemployeesofanorganisation.Theimpactofmaternalandchildhealthservicesontheinfantmortalityrate.Theeffectofanadvertisementonthesaleofaproduct.
Studydesignsbasedonthenatureoftheinvestigation
Onthebasisofthenatureoftheinvestigation,studydesignsinquantitativeresearchcanbeclassifiedas:
experimental;non-experimental;quasi-orsemi-experimental.
To understand the differences, let us consider some examples. Suppose you want to test thefollowing: the impactof aparticular teachingmethodon the levelof comprehensionof students; theeffectiveness of a programme such as random breath testing on the level of road accidents; or theusefulnessofadrugsuchasazidothymidine(AZT)intreatingpeoplewhoareHIV-positive;orimagineanysimilarsituationinyourownacademicorprofessionalfield.Insuchsituationsthereisassumedtobeacause-and-effectrelationship.Therearetwowaysofstudyingthisrelationship.Thefirstinvolvesthe researcher (or someone else) introducing the intervention that is assumed to be the ‘cause’ ofchange,andwaitinguntilithasproduced–orhasbeengivensufficienttimetoproduce–thechange.Thesecondconsistsoftheresearcherobservingaphenomenonandattemptingtoestablishwhatcausedit. In this instance the researcher starts from the effect(s) or outcome(s) and attempts to determinecausation.Ifarelationshipisstudiedinthefirstway,startingfromthecausetoestablishtheeffects,itisclassifiedasanexperimentalstudy.Ifthesecondpathisfollowed–thatis,startingfromtheeffectstotracethecause–itisclassifiedasanon-experimentalstudy(seeFigure8.6).
FIGURE8.6Experimentalandnon-experimentalstudiesIntheformercasetheindependentvariablecanbe‘observed’,introduced,controlledormanipulated
bytheresearcherorsomeoneelse,whereas in the latter thiscannothappenas theassumedcausehasalreadyoccurred. Instead, theresearcher retrospectively links thecause(s) to theoutcome(s).Asemi-experimental study orquasi-experimental study has the properties of both experimental and non-experimentalstudies;partofthestudymaybenon-experimentalandtheotherpartexperimental.Anexperimentalstudycanbecarriedoutineithera‘controlled’ora‘natural’environment.Foran
experimentinacontrolledenvironment,theresearcher(orsomeoneelse)introducestheinterventionorstimulustostudyitseffects.Thestudypopulationisina‘controlled’situationsuchasaroom.Foranexperiment in a ‘natural’ environment, the study population is exposed to an intervention in its ownenvironment.Experimentalstudiescanbefurtherclassifiedonthebasisofwhetherornotthestudypopulationis
randomlyassigned todifferent treatmentgroups.Oneof thebiggestproblems in comparabledesigns(thoseinwhichyoucomparetwoormoregroups)isalackofcertaintythatthedifferentgroupsarein
fact comparable in every respect except the treatment. The process of randomisation is designed toensure that the groups are comparable. In a randomdesign, the study population, the experimentaltreatmentsorbotharenotpredeterminedbutrandomlyassigned(seeFigure8.7).Randomassignmentin experiments means that any individual or unit of a study population group has an equal andindependent chanceofbecomingpartof anexperimentalor controlgroupor, in thecaseofmultipletreatmentmodalities,anytreatmenthasanequalandindependentchanceofbeingassignedtoanyofthepopulationgroups.Itisimportanttonotethattheconceptofrandomisationcanbeappliedtoanyoftheexperimentaldesignswediscuss.
FIGURE8.7Randomisationinexperiments
Experimentalstudydesigns
Therearesomanytypesofexperimentaldesignthatnotallofthemcanbeconsideredwithinthescopeofthisbook.Thissection,therefore,isconfinedtodescribingthosemostcommonlyusedinthesocialsciences, the humanities, public health,marketing, education, epidemiology, socialwork, and so on.Thesedesignshavebeencategorisedas:
theafter-onlyexperimentaldesign;thebefore-and-afterexperimentaldesign;thecontrolgroupdesign;thedouble-controldesign;thecomparativedesign;the‘matchedcontrol’experimentaldesign;theplacebodesign.
FIGURE8.8Theafter-onlydesign
Theafter-onlyexperimentaldesignIn an after-only design the researcher knows that a population is being, or has been, exposed to aninterventionandwishes tostudy its impacton thepopulation. In thisdesign, informationonbaseline(pre-test or before observation) is usually ‘constructed’ on the basis of respondents’ recall of thesituationbeforetheintervention,orfrominformationavailableinexistingrecords–secondarysources(Figure8.8).Thechangeinthedependentvariableismeasuredbythedifferencebetweenthe‘before’(baseline)and‘after’datasets.Technically,thisisaveryfaultydesignformeasuringtheimpactofaninterventionastherearenoproperbaselinedatatocomparethe‘after’observationwith.Therefore,oneof the major problems of this design is that the two sets of data are not strictly comparable. Forexample,someofthechangesinthedependentvariablemaybeattributabletothedifferenceinthewaythetwosetsofdatawerecompiled.Anotherproblemwiththisdesignisthatitmeasurestotalchange,including change attributable to extraneous variables; hence, it cannot identify the net effect of anintervention.However, this design iswidely used in impact assessment studies, as in real lifemanyprogrammes operate without the benefit of a planned evaluation at the programme planning stage(though this is fast changing) in which case it is just not possible to follow the sequence strictly –collectionofbaselineinformation,implementationoftheprogrammeandthenprogrammeevaluation.Anevaluatorthereforehasnochoicebuttoadoptthisdesign.Inpractice,theadequacyofthisdesigndependsonhavingreasonablyaccuratedataavailableabout
the prevalence of a phenomenon before the intervention is introduced. This might be the case forsituations such as the impact of random breath testing on road accidents, the impact of a healthprogrammeonthemortalityofapopulation,theimpactofanadvertisementonthesaleofaproduct,theimpactofadeclineinmortalityonthefertilityofapopulation,ortheimpactofachangeinimmigrationpolicy on the extent of immigration. In these situations it is expected that accurate records are keptaboutthephenomenonunderstudyandsoitmaybeeasiertodeterminewhetheranychangeintrendsisprimarilybecauseoftheintroductionoftheinterventionorchangeinthepolicy.
Thebefore-and-afterexperimentaldesignThe before-and-after design overcomes the problem of retrospectively constructing the ‘before’observationbyestablishingitbeforetheinterventionisintroducedtothestudypopulation(seeFigure8.2).Then,whentheprogrammehasbeencompletelyimplementedorisassumedtohavehaditseffecton the population, the ‘after’ observation is carried out to ascertain the impact attributable to the
intervention(seeFigure8.9).
FIGURE8.9Measurementofchangethroughabefore-and-afterdesignThe before-and-after design takes care of only one problem of the after-only design – that is, the
comparability of the before-and-after observations. It still does not enable one to conclude that anychange– inwholeor inpart– canbeattributed to theprogramme intervention.Toovercome this, a‘control’groupisused.Before-and-afterdesignsmayalsosufferfromtheproblemsidentifiedearlierinthis chapter in the discussion of before-and-after study designs. The impact of the intervention inbefore-and-afterdesigniscalculatedasfollows:
[changeindependentvariable]=
[statusofthedependentvariableatthe‘after’observation]–[statusofthedependentvariableatthe‘before’observation]
ThecontrolgroupdesignInastudyutilisingthecontrolgroupdesigntheresearcherselectstwopopulationgroupsinsteadofone:a control group and an experimental group (Figure 8.10). These groups are expected to becomparable as far as possible in every respect except for the intervention (that is assumed to be thecauseresponsibleforbringingaboutthechange).Theexperimentalgroupeitherreceivesorisexposedtotheintervention,whereasthecontrolgroupisnot.Firstly,the‘before’observationsaremadeonbothgroups at the same time. The experimental group is then exposed to the intervention. When it isassumed that the interventionhashadan impact,an ‘after’observation ismadeonbothgroups.Anydifference in the ‘before’ and ‘after’ observations between the groups regarding the dependentvariable(s)isattributedtotheintervention.
FIGURE8.10Thecontrolexperimentaldesign
Intheexperimentalgroup,totalchangeinthedependentvariable(Ye)canbecalculatedasfollows:
Ye=(Y″e–Y'e)
where
Y″e=‘after’observationontheexperimentalgroupY'e=‘before’observationontheexperimentalgroup
Inotherwords,
(Y″e–Y'e)= (impactofprogrammeintervention)±(impactofextraneousvariables)±(impactofchancevariables)
Inthecontrolgroup,totalchangeinthedependentvariable(Yc)canbecalculatedasfollows:
Yc=(Y″c–Y'c)
where
Y″c=post-testobservationonthecontrolgroupY'c=pre-testobservationonthecontrolgroup
Inotherwords,
(Y″c–Y'c)= (impactofextraneousvariables)±(impactofchancevariables)
Thedifferencebetweenthecontrolandexperimentalgroupscanbecalculatedas
(Y″e–Y'e)–(Y″c–Y'c),
whichis
{(impactofprogrammeintervention)±(impactofextraneousvariablesinexperimentalgroups)±(impactofchancevariablesinexperimentalgroups)}-{(impactofextraneousvariablesincontrolgroup)±(impactofchancevariablesincontrolgroup)}
Usingsimplearithmeticoperations,thisequalstheimpactoftheintervention.
Therefore, the impact of any intervention is equal to the difference in the ‘before’ and ‘after’observationsinthedependentvariablebetweentheexperimentalandcontrolgroups.
Itisimportanttorememberthatthechiefobjectiveofthecontrolgroupistoquantifytheimpactofextraneousvariables.Thishelpsyoutoascertaintheimpactoftheinterventiononly.
Thedouble-controldesignAlthough the control design helps you to quantify the impact that can be attributed to extraneousvariables,itdoesnotseparateoutothereffectsthatmaybeduetotheresearchinstrument(suchasthereactiveeffect)or respondents(suchas thematurationorregressioneffects,orplaceboeffect).Whenyouneedtoidentifyandseparateouttheseeffects,adouble-controldesignisrequired.Indouble-controlstudies,youhavetwocontrolgroupsinsteadofone.Toquantify,say,thereactive
effect of an instrument, you excludeoneof the control groups from the ‘before’ observation (Figure8.11).
FIGURE8.11Double-controldesigns
Youcancalculatethedifferenteffectsasfollows:
(Y″e–Y'e)=(impactofprogrammeintervention)±(impactofextraneousvariables)±(reactiveeffect)±(randomeffect)(Y″c1–Y'c1)=(impactofextraneousvariables)±(reactiveeffect)±(randomeffect)(Y″c2–Y'c1)=(impactofextraneousvariables)±(randomeffect)
(Notethat(Y″c2–Y'c1)andnot(Y″c2–Y'c2)asthereisno‘before’observationforthesecondcontrolgroup.)
(Y'e–Y'e)–(Y″c1–Y'c1)=impactofprogrammeintervention(Y″c1–Y'c1)–(Y'c2–Y'c1)=reactiveeffect
Theneteffectof theprogrammeinterventioncanbecalculatedinthesamemannerasforthecontrolgroupdesignsasexplainedearlier.
Thecomparativedesign
Sometimes you seek to compare the effectiveness of different treatment modalities and in suchsituationsacomparativedesignisappropriate.With a comparative design, as with most other designs, a study can be carried out either as an
experiment or as a non-experiment. In the comparative experimental design, the study population isdividedintothesamenumberofgroupsas thenumberof treatments tobetested.Foreachgroupthebaselinewithrespecttothedependentvariableisestablished.Thedifferenttreatmentmodelsarethenintroducedtothedifferentgroups.Afteracertainperiod,whenitisassumedthatthetreatmentmodelshave had their effect, the ‘after’ observation is carried out to ascertain any change in the dependentvariable. The degree of change in the dependent variable in the different population groups is thencomparedtoestablishtherelativeeffectivenessofthevariousinterventions.Inthenon-experimentalformofcomparativedesign,groupsalreadyreceivingdifferentinterventions
are identified,andonly thepost-observationwithrespect to thedependentvariable isconducted.Thepre-testdatasetisconstructedeitherbyaskingthestudypopulationineachgrouptorecalltherequiredinformation relating to the period before the introduction of the treatment, or by extracting suchinformation from existing records. Sometimes a pre-test observation is not constructed at all, on theassumptionthatifthegroupsarecomparablethebaselinemustbeidentical.Aseachgroupisassumedtohave thesamebaseline, thedifference in thepost-testobservation isassumed tobebecauseof theintervention.Toillustratethis,imagineyouwanttocomparetheeffectivenessofthreeteachingmodels(A,Band
C)onthelevelofcomprehensionofstudentsinaclass(Figure8.12).Toundertakethestudy,youdividetheclassintothreegroups(X,YandZ),throughrandomisation,toensuretheircomparability.Beforeexposingthesegroupstotheteachingmodels,youfirstestablishthebaselineforeachgroup’slevelofcomprehensionofthechosensubject.Youthenexposeeachgrouptoadifferentteachingmodeltoteachthechosensubject.Afterwards,youagainmeasurethegroups’levelsofcomprehensionofthematerial.SupposeXaistheaveragelevelofcomprehensionofgroupXbeforethematerialistaught,andXa'isthis group’s average level of comprehension after thematerial is taught. The change in the level ofcomprehension,Xa'–XaisthereforeattributedtomodelA.Similarly,changesingroupYandZ,Yb'–YbandZc'–Zc, are attributed to teachingmodelsB andC respectively.The changes in the averagelevelofcomprehensionforthethreegroupsarethencomparedtoestablishwhichteachingmodelisthemost effective. (Note that extraneous variables will affect the level of comprehension in all groupsequally,astheyhavebeenformedrandomly.)
FIGURE8.12ComparativeexperimentaldesignItisalsopossibletosetupthisstudyasanon-experimentalone,simplybyexposingeachgroupto
oneofthethreeteachingmodels,followingupwithan‘after’observation.Thedifferenceinthelevelsofcomprehensionisattributedto thedifferencein theteachingmodelsas it isassumedthat thethreegroupsarecomparablewithrespecttotheiroriginallevelofcomprehensionofthetopic.
ThematchedcontrolexperimentaldesignComparative groups are usually formed on the basis of their overall comparabilitywith respect to arelevantcharacteristicinthestudypopulation,suchassocioeconomicstatus,theprevalenceofacertaincondition or the extent of a problem in the study population. In matched studies, comparability isdeterminedon an individual-by-individual basis.Two individuals from the study populationwho arealmostidenticalwithrespecttoaselectedcharacteristicand/orcondition,suchasage,genderortypeofillness,arematchedandtheneachisallocatedtoaseparategroup(thematchingisusuallydoneonaneasilyidentifiablecharacteristic).Inthecaseofamatchedcontrolexperiment,oncethetwogroupsareformed,youasaresearcherdecidethroughrandomisationorotherwisewhichgroupistobeconsideredcontrol,andwhichexperimental.Thematcheddesigncanposeanumberofchallenges:
Matchingincreasesindifficultywhencarriedoutonmorethanonevariable.Matchingonvariablesthatarehardtomeasure,suchasattitudeoropinion,isextremelydifficult.Sometimesitishardtoknowwhichvariabletochooseasabasisformatching.Youmaybeabletobaseyourdecisionuponpreviousfindingsoryoumayhavetoundertakeapreliminarystudytodetermineyourchoiceofvariable.
Matchedgroupsaremostcommonlyusedinthetestingofnewdrugs.
The‘placebo’designApatient’sbeliefthats/heisreceivingtreatmentcanplayanimportantroleinhis/herrecoveryfromanillness even if treatment is ineffective. This psychological effect is known as theplacebo effect. Aplacebodesignattempts todeterminetheextentof thiseffect.Aplacebostudy involves twoor threegroups,dependingonwhetherornottheresearcherwantstohaveacontrolgroup(Figure8.13).Iftheresearcherdecides tohaveacontrolgroup, thefirstgroupreceives the treatment, thesecondreceivestheplacebo treatment and the third– thecontrolgroup– receivesnothing.Thedecisionas towhichgroupwillbethetreatment,theplaceboorthecontrolgroupcanalsobemadethroughrandomisation.
FIGURE8.13Theplacebodesign
Otherdesignscommonlyusedinquantitativeresearch
Therearesomeresearchdesignsthatmaybeclassifiedinthetypologydescribedabovebut,becauseoftheir uniqueness and prevalence, have acquired their own names. They are therefore describedseparatelybelow.
Thecross-overcomparativeexperimentaldesign
Thedenialoftreatmenttothecontrolgroupisconsideredunethicalbysomeprofessionals.Inaddition,thedenialoftreatmentmaybeunacceptabletosomeindividualsinthecontrolgroup,whichcouldresultin them dropping out of the experiment and/or going elsewhere to receive treatment. The formerincreases ‘experimental mortality’ and the latter may contaminate the study. The cross-overcomparative experimental design makes it possible to measure the impact of a treatment withoutdenyingtreatmenttoanygroup,thoughthisdesignhasitsownproblems.Inthecross-overdesign,alsocalledtheABABdesign(Grinnell1993:104),twogroupsareformed,
theinterventionisintroducedtooneofthemand,afteracertainperiod,theimpactofthisinterventionismeasured.Thentheinterventionsare‘crossedover’;thatis,theexperimentalgroupbecomesthecontroland vice versa, sometimes repeatedly over the period of the study (Figure 8.14). However, in thisdesign, population groups do not constitute experimental or control groups but only segments uponwhichexperimentalandcontrolobservationsareconducted.
FIGURE8.14Thecross-overexperimentaldesignOneofthemaindisadvantagesofthisdesignisdiscontinuityintreatment.Themainquestionis:what
impactwouldinterventionhaveproducedhaditnotbeenprovidedinsegments?
Thereplicatedcross-sectionaldesign
In practice one usually examines programmes already in existence and ones in which clients are atdifferent stages of an intervention. Evaluating the effectiveness of such programmes within aconventional experimental design is impossible because a baseline cannot be established as theintervention has already been introduced. In this situation, the usualmethod of selecting a group of
peoplewhowererecentlyrecruitedtotheprogrammeandfollowingthemthroughuntiltheinterventionhasbeencompletedmaytakealongtime.Insuchsituations,itispossibletochooseclientswhoareatdifferentphasesoftheprogrammetoformthebasisofyourstudy(Figure8.15).
FIGURE8.15Thereplicatedcross-sectionaldesignThis design is based upon the assumption that participants at different stages of a programme are
similarintermsoftheirsocioeconomic–demographiccharacteristicsandtheproblemforwhichtheyareseekingintervention.Assessmentoftheeffectivenessofaninterventionisdonebytakingasampleofclientsatdifferentstagesoftheintervention.Thedifferenceinthedependentvariableamongclientsatintakeandterminationstageisconsideredtobetheimpactoftheintervention.
Trendstudies
Ifyouwanttomapchangeoveraperiod,atrendstudyisthemostappropriatemethodofinvestigation.Trendanalysisenablesyoutofindoutwhathashappenedinthepast,whatishappeningnowandwhatislikelytohappeninthefutureinapopulationgroup.Thisdesigninvolvesselectinganumberofdataobservationpointsinthepast,togetherwithapictureofthepresentorimmediatepastwithrespecttothephenomenonunderstudy,andthenmakingcertainassumptionsastofuturetrends.Inawayyouarecollectingcross-sectionalobservationsabout the trendbeingobservedatdifferentpoints in timeoverpast–present–future.Fromthesecross-sectionalobservationsyoudrawconclusionsaboutthepatternofchange.Trendstudiesareusefulinmakingforecastingbyextrapolatingpresentandpasttrendsthusmakinga
valuablecontributiontoplanning.Trendsregardingthephenomenonunderstudycanbecorrelatedwithother characteristics of the study population. For example, youmaywant to examine the changes inpoliticalpreferenceofastudypopulationinrelationtoage,gender,incomeorethnicity.Thisdesigncanalsobeclassifiedasretrospective–prospectivestudyonthebasisofthereferenceperiodclassificationsystemdevelopedearlierinthischapter.
Cohortstudies
Cohort studies are based upon the existence of a common characteristic such as year of birth,graduationormarriage,withinasubgroupofapopulation.Supposeyouwanttostudytheemploymentpattern of a batch of accountants who graduated from a university in 1975, or study the fertilitybehaviour of women who weremarried in 1930. To study the accountants’ career paths you wouldcontact all the accountantswho graduated from the university in 1975 to find out their employmenthistories.Similarly,youwould investigate the fertilityhistoryof thosewomenwhomarried in1930.Bothofthesestudiescouldbecarriedouteitherascross-sectionalorlongitudinaldesigns.Ifyouadopta cross-sectional design you gather the required information in one go, but if you choose thelongitudinaldesignyoucollecttherequiredinformationatdifferentpointsintimeoverthestudyperiod.Boththesedesignshavetheirstrengthsandweaknesses. In thecaseofa longitudinaldesign, it isnotimportant for the required information to be collected from the same respondents; however, it isimportantthatalltherespondentsbelongtothecohortbeingstudied;thatis,intheaboveexamplestheymusthavegraduatedin1975ormarriedin1930.
Panelstudies
Panelstudiesaresimilartotrendandcohortstudiesexceptthatinadditiontobeinglongitudinaltheyarealsoprospective innatureand the information isalwayscollectedfromthesamerespondents. (Intrend and cohort studies the information can be collected in a cross-sectional manner and theobservationpointscanbe retrospectivelyconstructed.)Supposeyouwant tostudy thechanges in thepatternofexpenditureonhouseholditemsinacommunity.Todothis,youwouldselectafewfamiliestofindouttheamounttheyspendeveryfortnightonhouseholditems.Youwouldkeepcollectingthesame information from the same families over a period of time to ascertain the changes in theexpenditurepattern.Similarly,apanelstudydesigncouldbeused tostudy themorbiditypattern inacommunity.
Blindstudies
Theconceptofablindstudy canbeusedwithcomparableandplaceboexperimentaldesignsand isappliedtostudiesmeasuringtheeffectivenessofadrug.Inablindstudy,thestudypopulationdoesnotknowwhether it isgetting realor fake treatmentorwhich treatmentmodality.Themainobjectiveofdesigningablindstudyistoisolatetheplaceboeffect.
Double-blindstudies
Theconceptofadouble-blindstudyisverysimilartothatofablindstudyexceptthatitalsotriestoeliminate researcherbiasbyconcealing the identityof theexperimentalandplacebogroups from theresearcher. In other words, in a double-blind study neither the researcher nor the study participantsknowwho is receiving real andwho is receiving fake treatment orwhich treatmentmodel they arereceiving.
Studydesignsinqualitativeresearch
Thissectionprovidesabriefdescriptionofsomeofthecommonlyuseddesignsinqualitativeresearch.
Foranin-depthunderstandingyouareadvisedtoconsultbooksonqualitativeresearch.
Casestudy
Thecasestudy,thoughdominantlyaqualitativestudydesign,isalsoprevalentinquantitativeresearch.Acasecouldbeanindividual,agroup,acommunity,aninstance,anepisode,anevent,asubgroupofapopulation,atownoracity.Tobecalledacasestudyitisimportanttotreatthetotalstudypopulationasoneentity.Inacasestudydesign the ‘case’youselectbecomes thebasisofa thorough,holisticand in-depth
exploration of the aspect(s) that youwant to find out about. It is an approach ‘inwhich a particularinstance or a few carefully selected cases are studied intensively’ (Gilbert 2008: 36). According toBurns(1997:364),‘toqualifyasacasestudy,itmustbeaboundedsystem,anentityinitself.Acasestudyshouldfocusonaboundedsubject/unitthatiseitherveryrepresentativeorextremelyatypical.’Acase study according to Grinnell (1981: 302), ‘is characterized by a very flexible and open-endedtechniqueofdatacollectionandanalysis’.Thecasestudydesignisbasedupontheassumptionthatthecasebeingstudiedisatypicalofcasesof
acertaintypeandthereforeasinglecasecanprovideinsightintotheeventsandsituationsprevalentinagroupfromwherethecasehasbeendrawn.AccordingtoBurns(1997:365),‘Inacasestudythefocusof attention is the case in its idiosyncratic complexity, not on the whole population of cases.’ Inselecting a case therefore you usually use purposive, judgemental or information-oriented samplingtechniques.Itisaveryusefuldesignwhenexploringanareawherelittleisknownorwhereyouwanttohavea
holisticunderstandingofthesituation,phenomenon,episode,site,grouporcommunity.Thisdesignisofimmenserelevancewhenthefocusofastudyisonextensivelyexploringandunderstandingratherthan confirming and quantifying. It provides an overview and in-depth understanding of a case(s),processandinteractionaldynamicswithinaunitofstudybutcannotclaimtomakeanygeneralisationstoapopulationbeyondcasessimilartotheonestudied.Inthisdesignyourattemptisnottoselectarandomsamplebutacasethatcanprovideyouwithas
much information as possible to understand the case in its totality.When studying an episode or aninstance, you attempt to gather information from all available sources so as to understand it in itsentirety.Ifthefocusofyourstudyisagrouporcommunityyoushouldspendsufficienttimebuildingatrustworthyrapportwithitsmembersbeforecollectinganyinformationaboutthem.Though you can use a singlemethod, the use ofmultiplemethods to collect data is an important
aspect of a case study, namely in-depth interviewing, obtaining information from secondary records,gatheringdatathroughobservations,collectinginformationthroughfocusgroupsandgroupinterviews,etc.However, it is importantthatat thetimeofanalysisyoucontinuetoconsiderthecaseasasingleentity.
Oralhistory
Oralhistoryismoreamethodofdatacollectionthanastudydesign;however,inqualitativeresearch,thishasbecomeanapproach tostudyperceptions,experiencesandaccountsofaneventorgatheringhistoricalknowledgeasviewedby individuals. It isapictureof something insomeone’sownwords.Oral history is a process of obtaining, recording, presenting and interpreting historical or currentinformation,baseduponpersonalexperiencesandopinionsofsomemembersofastudygrouporunit.Theseopinionsorexperiencescouldbebaseduponeye-witnessevidenceorinformationpassedonfrom
other sources such as older people, ancestors, folklore, stories. According to Ritchie (2003: 19),‘Memoryisthecoreoforalhistory,fromwhichmeaningcanbeextractedandpreserved.Simplyput,oralhistorycollectsmemories andpersonal commentariesofhistorical significance through recordedinterviews.’AccordingtoBurns(1997:368),‘theseareusuallyfirstpersonnarrativesthattheresearchercollectsusingextensiveinterviewingofasingleindividual’.Intermsofdesignitisquitesimple.Youfirstdecidewhattypesofaccount,experience,perceptionor
historicaleventyouwanttofindoutabout.Thenyouneedtoidentifytheindividualsorsources(whichcouldbedifficultandtimeconsuming)thatcanbestprovideyouwiththeneededinformation.Youthencollectinformationfromthemtobeanalysedandinterpreted.
Focusgroups/groupinterviews
Focusgroupsareaformofstrategyinqualitativeresearchinwhichattitudes,opinionsorperceptionstowards an issue, product, service or programme are explored through a free and open discussionbetweenmembersofagroupandtheresearcher.Bothfocusgroupsandgroupinterviewsarefacilitatedgroupdiscussionsinwhicharesearcherraisesissuesorasksquestionsthatstimulatediscussionamongmembersofthegroup.Becauseofitslowcost,itisapopularmethodforfindinginformationinalmosteveryprofessionalareaandacademicfield.Social,politicalandbehaviouralscientists,marketresearchandproducttestingagencies,andurbanandtownplanningexpertsoftenusethisdesignforavarietyofsituations.Forexample,inmarketingresearchthisdesigniswidelyusedtofindoutconsumers’opinionofandfeedbackonaproduct, theiropinionson thequalityof theproduct, itsacceptanceandappeal,priceandpackaging,howtoimprovethequalityandincreasethesaleoftheproduct,etc.Focusgroupsarealsoprevalentinformativeandsummativeevaluationsandfordevelopingsocialprogrammesandservices.Itisalsoausefultoolinsocialandurbanplanningforidentifyingissues,options,developmentstrategies,andfutureplanninganddevelopmentdirections.Initsdesignit isverysimple.Youasaresearcherselectagroupofpeoplewhoyouthinkarebest
equipped to discusswhat youwant to explore. The group could comprise individuals drawn from agroup of highly trained professionals or average residents of a community depending upon theobjectivesof thefocusgroup.In theformationofafocusgroupthesizeof thegroupisanimportantconsideration.Itshouldbeneithertoolargenortoosmallasthiscanimpedeupontheextentandqualityofthediscussion.Approximatelyeighttotenpeoplearetheoptimalnumberforsuchdiscussiongroups.Youalsoneedtoidentifycarefullytheissuesfordiscussionprovidingeveryopportunityforadditionalrelevantones toemerge.Asa researcheryoualsoneed todecide, inconsultationwith thegroup, theprocess of recording the discussion. This may include fixing the times that the group can meet toextensivelydiscussingthe issuesandarrivingatagreementsonthem.Yourrecordsof thediscussionsthenbecome thebasisofanalysis for findingsandconclusions.Themaindifferencebetweena focusgroupandagroupinterviewisinthedegreeofspecificitywithrespecttotheissuestobediscussed.Theissuesdiscussed in focusgroupsaremorespecificandfocused than ingroup interviewsand theyarelargely predetermined by the researcher. In a group interview you let the group members discusswhatevertheywant.However,yourroleasaresearcheristobringthembacktotheissuesofinterestasidentifiedbythegroup.Compared with other designs this is less expensive and needs far less time to complete. The
information generated can be detailed and rich and can be used to explore a vast variety of issues.However,thedisadvantageisthatifthediscussionisnotcarefullydirecteditmayreflecttheopinionofthosewhohaveatendencytodominateagroup.Thisdesignisveryusefulforexploringthediversityinopinionsondifferentissuesbutwillnothelpyouifyouwanttofindouttheextentormagnitudeofthis
diversity.
Participantobservation
Participantobservation is another strategy for gathering information about a social interaction or aphenomenoninqualitativestudies.Thisisusuallydonebydevelopingacloseinteractionwithmembersof a group or ‘living’ in the situation which is being studied. Though predominantly a qualitativeresearchdesign,itisalsousedinquantitativeresearch,dependinguponhowtheinformationhasbeengenerated and recorded. In qualitative research, an observation is always recorded in a descriptiveformatwhereasinquantitativeresearchitisrecordedeitherincategoriesoronascale.Itcanalsobeacombinationofboth– somecategorisationand somedescriptionor categorisationaccompaniedbyadescriptive explanation. You can also change a descriptive recording into a categorical one throughanalysis and classification. In addition to the observation itself, where you as an observer generateinformation,theinformationcanalsobecollectedthroughothermethodssuchasinformalinterviewing,in-depthinterviewing,groupdiscussions,previousdocuments,oralhistories.Useofmultiplemethodswillenhancetherichnessoftheinformationcollectedbyparticipantobservation.In its design it is simple.You as a researcher get involved in the activities of the group, create a
rapport with group members and then, having sought their consent, keenly observe the situation,interaction,siteorphenomenon.Youmakedetailednotesofwhatyouobserveinaformatthatbestsuitsyouaswellasthesituation.Youcanalsocollectinformationusingothermethodsofdatacollection,ifneedbe.Youanalyserecordsofyourobservationsanddatacollectedbyothermeanstodrawinferencesandconclusions.Themainadvantageofparticipantobservationisthatasyouspendsufficienttimewiththegroupor
inthesituation,yougainmuchdeeper,richerandmoreaccurateinformation,butthemaindisadvantageisthat,ifyouarenotverycareful,youcanintroduceyourownbias.
Holisticresearch
Theholisticapproachtoresearchisonceagainmoreaphilosophythanastudydesign.Thedesignisbaseduponthephilosophythatasamultiplicityoffactorsinteractsinourlives,wecannotunderstandaphenomenonfromjustoneortwoperspectives.Tounderstandasituationorphenomenonyouneedtolookatitinitstotality–thatis,holisticallyfromeveryperspective.Youcanuseanydesignwhenexploringasituationfromdifferentperspectivesandtheuseofmultiple
methodsisprevalentanddesirable.
Communitydiscussionforums
Communitydiscussionforumsaredesignedtofindopinions,attitudesand/orideasofacommunitywithregardtocommunityissuesandproblems.Itisoneoftheverypopularwaysofseekingacommunity’sparticipationindecidingaboutissuesofconcerntomembersofthecommunity.Suchforumsarealsoused for a variety of other reasons such as developing townplanning options and community healthprogrammesforacommunity,seekingparticipationofitsmembersinresolvingissuesrelatingtotrafficmanagement, infrastructure development and determining future directions for the area, informingcommunitiesofnewinitiatives.Communityforumsareverysimilar togroupdiscussionsexcept that theseareonabiggerscale in
terms of number of participants. Also, in group discussions youmay select the participants, but forcommunity forums there is self-selection of the participants as they are open to everyone with aninterestintheissuesorconcerns.Theresearcherusuallyuseslocalmediatoinformtheresidentsofalocalcommunityabouttheforums.This is a useful design to find out the spread of issues, concerns, etc., at a community level. It is
economicalandquickbuttherearesomedisadvantages.Forexample, it ispossiblethatafewpeoplewithavestedinterestcandominatethediscussioninaforumanditisequallypossiblethatonoccasionsthere may be very low attendance. Such situations may result in the discussion not reflecting thecommunityattitudes.
Reflectivejournallog
Basically,thisdesignentailskeepingareflectivejournallogofyourthoughtsasaresearcherwheneveryou notice anything, talk to someone, participate in an activity or observe something that helps youunderstandoraddtowhateveryouaretryingtofindoutabout.Thesereflectiverecordsthenbecomethebasisofyour findingsandconclusions.Youcanhavea reflective journalas theonlymethodofdatacollectionoritcanbeusedincombinationwithothermethodssuchasinterviewing,groupinterviews,orsecondarysources.
Othercommonlyusedphilosophy-guideddesigns
Thereareanumberofotherapproachestoresearchthathaveacquiredrecognition,intermsofdesignandname,intheresearchliterature.Whilenotdesignsperse,theydoenhanceaparticularphilosophicalperspectiveinsocialresearch.Theseare:actionresearch,feministresearch,participatoryresearchandcollaborative enquiry. Strictly speaking, a piece of research within each of these could be eitherquantitativeorqualitative,thoughbymanytheyareconsidereddominantlyasqualitativedesigns.Theneed to place them in a separate category stems from their prominence and possible use in eachparadigm.Thesedesignsaremorephilosophyguidedthanmethodsbased.Forexample,actionresearchisguidedby thephilosophy thatapieceof researchshouldbefollowedbysomeformofappropriateactiontoachievebettermentinlifeorservice,andfeministresearchisinfluencedbythephilosophythatopposesandchallengesthedominantmalebiasinsocialscienceresearch;itseemstobelievethatissuesrelatingtowomenarebestunderstoodandresearchedbywomenalone.Forparticipatoryresearchandcollaborativeenquiry,theinvolvementofresearchparticipantsorthecommunityintheresearchprocessistheunderlyingphilosophy.Oneoftheimportantaspectsofallthese‘designs’isthattheyattempttoinvolveresearchparticipantsintheresearchprocess.Theresearchfindingsarethenusedtodepictthecurrentsituationwithrespecttocertainissuesorproblemsandhelptoformasoundbasisforstrategydevelopmenttodealwiththem.
Actionresearch
As the name suggests,action research comprises two components: action and research (see Figure8.16).Research is ameans toaction, either to improveyourpracticeor to takeaction todealwithaproblemoranissue.Sinceactionresearchisguidedbythedesiretotakeaction,strictlyspeakingitisnotadesignperse.Mostactionresearchisconcernedwithimprovingthequalityofservice.Itiscarriedouttoidentifyareasofconcern,developandtestalternatives,andexperimentwithnewapproaches.
FIGURE8.16ActionresearchdesignActionresearchseemstofollowtwotraditions.TheBritishtraditiontendstoviewactionresearchas
a means of improvement and advancement of practice (Carr & Kemmis 1986), whereas in the UStraditionitisaimedatsystematiccollectionofdatathatprovidesthebasisforsocialchange(Bogdan&Biklen1992).Actionresearch,incommonwithparticipatoryresearchandcollaborativeenquiry,isbasedupon
a philosophy of community development that seeks the involvement of community members.Involvement and participation of a community, in the total process from problem identification toimplementation of solutions, are the two salient features of all three approaches (action research,participatory research and collaborative enquiry). In all three, data is collected through a researchprocess, and changes are achieved through action. This action is taken either by officials of aninstitutionorthecommunityitselfinthecaseofactionresearch,orbymembersofacommunityinthecaseofcollaborativeorparticipatoryresearch.Therearetwofocusesofactionresearch:
1. Anexistingprogrammeorinterventionisstudiedinordertoidentifypossibleareasofimprovementintermsofenhancedefficacyand/orefficiency.Thefindingsbecomethebasisofbringingaboutchanges.
2. Aprofessionalidentifiesanunattendedproblemorunexplainedissueinthecommunityoramongaclientgroupandresearchevidenceisgatheredtojustifytheintroductionofanewserviceorintervention.Researchtechniquesestablishtheprevalenceoftheproblemortheimportanceofanissuesothatappropriateactioncanbetakentodealwithit.
Feministresearch
Feministresearchischaracterisedbyitsfeministtheoryphilosophicalbasethatunderpinsallenquiriesandfeministconcernsactastheguidingframework.Feministresearchdiffersfromtraditionalresearchinthreeways:
1. Itsmainfocusistheexperiencesandviewpointsofwomen.Itusesresearchmethodsaimedatexploringthese.
2. Itactivelytriestoremoveorreducethepowerimbalancebetweentheresearcherandrespondents.
3. Thegoaloffeministresearchischangingthesocialinequalitybetweenmenandwomen.Infact,feministresearchmaybeclassifiedasactionresearchintheareaofgenderinequality,usingresearchtechniquestocreateawarenessofwomen’sissuesandconcerns,andtofosteractionpromotingequalitybetweensexes.
Anystudydesigncouldbeusedinfeministresearch.
Participatoryandcollaborativeresearchenquiry
As alreadymentioned, to the author’smind, these are not designs per se but signify a philosophicalperspective that advocates the active involvement of research participants in the research process.Participatoryresearchisbasedupontheprincipleofminimisingthe‘gap’betweentheresearcherandthe research participants and increased community involvement and participation to enhance therelevanceoftheresearchfindingstotheirneeds.Itisassumedthatsuchinvolvementwillincreasethepossibility of the community accepting the research findings and, if need be, its willingness andinvolvement in solving the problems and issues that confront it.You canundertake a quantitative orqualitativestudyintheseenquiriesbutthemainemphasisisonpeople’sengagement,collaborationandparticipationintheresearchprocess.Inawaythesedesignsarebasedonthecommunitydevelopmentmodelwhere engagement of a community byway of consultation and participation in planning andexecutionof research tasks is imperative. In thesedesignsyouarenotmerelya researcherbutalsoacommunityorganiserseekingactiveparticipationofthecommunity.As a researcher you work at two different aspects: (1) community organisation and (2) research.
Through community organisationyou seek a community’s involvement andparticipation in planningandexecutionoftheresearchtasksandshareresearchfindingswithitsmembers.Intermsofresearch,your main responsibility is to develop, in consultation with the community, the research tasks andprocedures.Consultationwithresearchparticipantsisacontinuousandintegralpartofthesedesigns.
SummaryInthischaptervariousstudydesignsinbothquantitativeandqualitativeresearchhavebeenexamined.Foreachstudydesign,detailshavebeenprovidedon the situations inwhich thedesign is appropriate touse, its strengthsandweaknesses, and theprocessyouadoptinitsoperationalisation.Inquantitativeresearchthevariousstudydesignshavebeenexaminedfromthreeperspectives.Theterminologyusedtodescribe
theseperspectives is thatof theauthorbut thenamesof thestudydesignsareuniversallyused.Thedifferentstudydesignsacrosseachcategoryaremutuallyexclusivebutnotsowithinacategory.The threeperspectives are thenumberof contacts, the referenceperiodand thenatureof the investigation.The first comprises
cross-sectional studies, before-and-after studies and longitudinal studies. The second categorises the studies as retrospective,prospective and retrospective–prospective. The third perspective classifies studies as experimental, non-experimental and semi-experimentalstudies.Qualitativestudydesignsarenotasspecific,preciseandwelldefinedasdesignsinquantitativeresearch.Also,thereisadegreeof
overlapbetweenstudydesignsandmethodsofdatacollection.Somedesignscaneasilybeconsideredasmethodsofdatacollection.Some of the commonly used designs in qualitative research are: case study design, oral history, focus group studies, participantobservation,communitydiscussionforumsandreflectivejournallog.Fouradditionalapproachestoresearchhavebeendescribed:actionresearch,feministresearch,andparticipatoryandcollaborative
enquiries.Though thesecannot reallybeconsidereddesigns in themselves, theyhaveacquired theirownidentity.Bothactionandfeminist research can be carried out either quantitatively or qualitatively, but participatory and collaborative enquiries are usuallyqualitativeinnature.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Identifytwoorthreesituationsrelatingtoyourownareaofinterestwhereyouthinkqualitativestudydesignsmightbemorebeneficialandconsiderwhythismightbethecase.Takeanexamplefromyourownacademicfieldorprofessionalareawhereanexperimental-controlorplacebogroupmightbeusedandexploretheethicalissuesrelatingtothis.
STEPIIIConstructinganInstrumentforDataCollection
Thisoperationalstepincludesthreechapters:
Chapter9:SelectingamethodofdatacollectionChapter10:CollectingdatausingattitudinalscalesChapter11:Establishingthevalidityandreliabilityofaresearchinstrument
CHAPTER9SelectingaMethodofDataCollection
Inthischapteryouwilllearnabout:
DifferencesinmethodsofdatacollectioninquantitativeandqualitativeresearchMajorapproachestoinformationgatheringCollectingdatausingprimarysources
ObservationTheinterviewThequestionnaire
MethodsofdatacollectioninqualitativeresearchCollectingdatausingsecondarysources
Keywords:closedquestions,contentanalysis,double-barrelledquestions,elevationeffect,errorofcentraltendency,focusgroup,haloeffect,Hawthorneeffect,interviewschedule,leadingquestions,non-participantobservation,open-endedquestions,oralhistory, participant observation, primary data, primary sources, questionnaire,secondarydata,secondarysources,structuredinterview,unstructuredinterview.
Differencesinthemethodsofdatacollectioninquantitativeandqualitativeresearch
Most methods of data collection can be used in both qualitative and quantitative research. Thedistinctionismainlyduetotherestrictionsimposedonflexibility,structure,sequentialorder,depthandfreedom that a researcher has in their use during the research process. Quantitativemethods favourthese restrictionswhereasqualitativeonesadvocateagainst them.Theclassificationofamethod intothequantitativeorqualitativecategorydependsuponyouranswerstothefollowingquestions:
Whatphilosophicalepistemologyisunderpinningyourapproachtoresearchenquiry?Howwastheinformationcollected?Wasitthroughastructuredorunstructured/flexibleformatof
datacollection?Werethequestionsorissuesdiscussedduringdatacollectionpredeterminedordevelopedduringdatacollection?Howwastheinformationyougatheredrecorded?Wasitinadescriptive,narrative,categorical,quantitativeformoronascale?Howwastheinformationanalysed?Wasitadescriptive,categoricalornumericalanalysis?Howdoyouproposetocommunicatethefindings?Doyouwanttowriteinadescriptiveoranalyticalmanner?
Forexample,ifanobservationisrecordedinanarrativeordescriptiveformat,itbecomesqualitativeinformation,but if it is recordedincategoricalformoronascale, itwillbeclassifiedasquantitativeinformation.Similarly fordatacollected through interviews.Anunstructured interview, recorded inadescriptive or narrative form, becomes a qualitative method, but in a structured interview, if theinformationisrecordedinresponsecategoriesorifthecategoriesaredevelopedandquantifiedoutofdescriptiveresponses,itisaquantitativemethod.Descriptiveresponsesobtainedinreplytoopen-endedquestionsareallqualitativebutiftheresponsesareinnumeralstheywillbeconsideredquantitative.Ifyoudevelopcategoriesandquantifythecategorisationasapartoftheanalysisofdescriptiveresponsesto an open-ended question, it becomes a quantitative analysis.Data generated by focus groups, oralhistories,narratives,groupinterviewsisalwaysqualitativeinnature.
Majorapproachestoinformationgathering
There are two major approaches to gathering information about a situation, person, problem orphenomenon.Whenyouundertakearesearchstudy,inmostsituations,youneedtocollecttherequiredinformation; however, sometimes the information required is already available and need only beextracted.Baseduponthesebroadapproachestoinformationgathering,datacanbecategorisedas:
primarydata;secondarydata.
FIGURE9.1Methodsofdatacollection
Informationgatheredusingthefirstapproachissaidtobecollectedfromprimarysources,whereas
thesourcesused in thesecondapproacharecalledsecondarysources.Examplesofprimary sourcesinclude finding out first-hand the attitudes of a community towards health services, ascertaining thehealth needs of a community, evaluating a social programme, determining the job satisfaction of theemployeesofanorganisation,andascertainingthequalityofserviceprovidedbyaworkerareexamplesof information collected from primary sources. On the other hand, the use of census data to obtaininformation on the age–sex structure of a population, the use of hospital records to find out themorbidityandmortalitypatternsofacommunity, theuseofanorganisation’s records toascertain itsactivities, and the collection of data from sources such as articles, journals, magazines, books andperiodicalstoobtainhistoricalandothertypesofinformation,areallclassifiedassecondarysources.Insummary,primarysourcesprovidefirst-handinformationandsecondarysourcesprovidesecond-handdata.Figure9.1showsthevariousmethodsofdatacollection.Noneofthemethodsofdatacollectionprovides100percentaccurateandreliableinformation.The
qualityofthedatagatheredisdependentuponanumberofotherfactors,whichwewillidentifyaswediscusseachmethod.Yourskillasaresearcherliesinyourabilitytotakecareofthefactorsthatcouldaffect the quality of your data. One of the main differences between experienced and amateurresearchersliesintheirunderstandingof,andabilitytocontrol,thesefactors.Itisthereforeimportantforabeginnertobeawareofthem.
Collectingdatausingprimarysources
Severalmethodscanbeusedtocollectprimarydata.Thechoiceofamethoddependsuponthepurposeofthestudy,theresourcesavailableandtheskillsoftheresearcher.Therearetimeswhenthemethodmostappropriate toachievetheobjectivesofastudycannotbeusedbecauseofconstraintssuchasalackof resources and/or required skills. In such situationsyou shouldbe awareof theproblems thattheselimitationsimposeonthequalityofthedata.Inselectingamethodofdatacollection,thesocioeconomic–demographiccharacteristicsofthestudy
populationplayanimportantrole:youshouldknowasmuchaspossibleaboutcharacteristicssuchaseducationallevel,agestructure,socioeconomicstatusandethnicbackground.Ifpossible,itishelpfultoknow the study population’s interest in, and attitude towards, participation in the study. Somepopulations, for a number of reasons, may not feel either at ease with a particular method of datacollection (such as being interviewed) or comfortable with expressing opinions in a questionnaire.Furthermore,peoplewithlittleeducationmayresponddifferentlytocertainmethodsofdatacollectioncomparedwithpeoplewithmoreeducation.Anotherimportantdeterminantofthequalityofyourdataisthewaythepurposeandrelevanceofthe
studyareexplainedtopotentialrespondents.Whatevermethodofdatacollectionisused,makesurethatrespondents clearly understand the purpose and relevanceof the study.This is particularly importantwhen you use a questionnaire to collect data, because in an interview situation you can answer arespondent’squestionsbutinaquestionnaireyouwillnothavethisopportunity.In the followingsectionseachmethodofdatacollection isdiscussed from thepointofviewof its
applicabilityandsuitabilitytoasituation,andtheproblemsandlimitationsassociatedwithit.
Observation
Observationisonewaytocollectprimarydata.Observationisapurposeful,systematicandselectiveway of watching and listening to an interaction or phenomenon as it takes place. There are manysituations inwhichobservation is themostappropriatemethodofdatacollection; forexample,whenyouwanttolearnabouttheinteractioninagroup,studythedietarypatternsofapopulation,ascertainthefunctionsperformedbyaworker,orstudythebehaviourorpersonalitytraitsofanindividual.Itisalsoappropriateinsituationswherefulland/oraccurateinformationcannotbeelicitedbyquestioning,becauserespondentseitherarenotco-operativeorareunawareoftheanswersbecauseitisdifficultforthem to detach themselves from the interaction. In summary, when you are more interested in thebehaviourthanintheperceptionsofindividuals,orwhensubjectsaresoinvolvedintheinteractionthattheyareunabletoprovideobjectiveinformationaboutit,observationisthebestapproachtocollecttherequiredinformation.
Typesofobservation
Therearetwotypesofobservation:
1. participantobservation;2. non-participantobservation.
Participantobservationiswhenyou,asaresearcher,participateintheactivitiesofthegroupbeingobserved in the same manner as its members, with or without their knowing that they are beingobserved. For example, youmight want to examine the reactions of the general population towardspeopleinwheelchairs.Youcanstudytheirreactionsbysittinginawheelchairyourself.Oryoumightwanttostudythelifeofprisonersandpretendtobeaprisonerinordertodothis.Non-participantobservation,ontheotherhand,iswhenyou,asaresearcher,donotgetinvolvedin
the activities of the group but remain a passive observer,watching and listening to its activities anddrawing conclusions from this. For example, you might want to study the functions carried out bynurses in a hospital. As an observer, you could watch, follow and record the activities as they areperformed.Aftermaking a number of observations, conclusions could be drawn about the functionsnurses carry out in the hospital.Anyoccupational group in any setting can be observed in the samemanner.
Problemswithusingobservationasamethodofdatacollection
Theuseofobservationasamethodofdatacollectionmaysufferfromanumberofproblems,whichisnottosuggestthatalloranyofthesenecessarilyprevailineverysituation.Butasabeginneryoushouldbeawareofthesepotentialproblems:
Whenindividualsorgroupsbecomeawarethattheyarebeingobserved,theymaychangetheirbehaviour.Dependinguponthesituation,thischangecouldbepositiveornegative–itmayincreaseordecrease,forexample,theirproductivity–andmayoccurforanumberofreasons.WhenachangeinthebehaviourofpersonsorgroupsisattributedtotheirbeingobserveditisknownastheHawthorneeffect.Theuseofobservationinsuchasituationmayintroducedistortion:whatisobservedmaynotrepresenttheirnormalbehaviour.
Thereisalwaysthepossibilityofobserverbias.Ifanobserverisnotimpartial,s/hecaneasilyintroducebiasandthereisnoeasywaytoverifytheobservationsandtheinferencesdrawnfromthem.Theinterpretationsdrawnfromobservationsmayvaryfromobservertoobserver.Thereisthepossibilityofincompleteobservationand/orrecording,whichvarieswiththemethodofrecording.Anobservermaywatchkeenlybutattheexpenseofdetailedrecording.Theoppositeproblemmayoccurwhentheobservertakesdetailednotesbutindoingsomissessomeoftheinteraction.
Situationsinwhichobservationscanbemade
Observationscanbemadeundertwoconditions:
1. natural;2. controlled.
Observing a group in its natural operation rather than intervening in its activities is classified asobservationundernaturalconditions.Introducingastimulustothegroupforittoreacttoandobservingthereactioniscalledcontrolledobservation.
Recordingobservations
Therearemanywaysofrecordingobservations.Theselectionofamethodofrecordingdependsuponthe purpose of the observation. Theway an observation is recorded also determines whether it is aquantitative or qualitative study. Narrative and descriptive recording is mainly used in qualitativeresearchbutifyouaredoingaquantitativestudyyouwouldrecordanobservationincategoricalformoronanumericalscale.Keepinmindthateachmethodofrecordinganobservationhasitsadvantagesanddisadvantages:
Narrativerecording–Inthisformofrecordingtheresearcherrecordsadescriptionoftheinteractioninhis/herownwords.Suchatypeofrecordingclearlyfallsinthedomainofqualitativeresearch.Usually,aresearchermakesbriefnoteswhileobservingtheinteractionandthensoonaftercompletingtheobservationmakesdetailednotesinnarrativeform.Inaddition,someresearchersmayinterprettheinteractionanddrawconclusionsfromit.Thebiggestadvantageofnarrativerecordingisthatitprovidesadeeperinsightintotheinteraction.However,adisadvantageisthatanobservermaybebiasedinhis/herobservationand,therefore,theinterpretationsandconclusionsdrawnfromtheobservationmayalsobebiased.Inaddition,interpretationsandconclusionsdrawnareboundtobesubjectivereflectingtheresearcher’sperspectives.Also,ifaresearcher’sattentionisonobserving,s/hemightforgettorecordanimportantpieceofinteractionand,obviously,intheprocessofrecording,partoftheinteractionmaybemissed.Hence,thereisalwaysthepossibilityofincompleterecordingand/orobservation.Inaddition,whentherearedifferentobserversthecomparabilityofnarrativerecordingcanbeaproblem.Usingscales–Attimessomeobserversmayprefertodevelopascaleinordertoratevariousaspectsoftheinteractionorphenomenon.Therecordingisdoneonascaledevelopedbythe
observer/researcher.Ascalemaybeone-,two-orthree-directional,dependinguponthepurposeoftheobservation.Forexample,inthescaleinFigure9.2–designedtorecordthenatureoftheinteractionwithinagroup–therearethreedirections:positive,negativeandneutral.Themainadvantageofusingscalesinrecordingobservationisthatyoudonotneedtospendtimeontakingdetailednotesandcanthusconcentrateonobservation.Ontheotherhand,theproblemswithusingascalearethatitdoesnotprovidespecificandin-depthinformationabouttheinteraction.Inaddition,itmaysufferfromanyofthefollowingerrors:
Unlesstheobserverisextremelyconfidentofhis/herabilitytoassessaninteraction,s/hemaytendtoavoidtheextremepositionsonthescale,usingmostlythecentralpart.Theerrorthatthistendencycreatesiscalledtheerrorofcentraltendency.Someobserversmayprefercertainsectionsofthescaleinthesamewaythatsometeachersarestrictmarkersandothersarenot.Whenobservershaveatendencytouseaparticularpartofthescaleinrecordinganinteraction,thisphenomenonisknownastheelevationeffect.Anothertypeoferrorthatmaybeintroducediswhenthewayanobserverratesanindividualononeaspectoftheinteractioninfluencestheways/heratesthatindividualonanotheraspectoftheinteraction.Againsomethingsimilartothiscanhappeninteachingwhenateacher’sassessmentoftheperformanceofastudentinonesubjectmayinfluencehis/herratingofthatstudent’sperformanceinanother.Thistypeofeffectisknownasthehaloeffect.
Categoricalrecording–Sometimesanobservermaydecidetorecordhis/herobservationusingcategories.Thetypeandnumberofcategoriesdependuponthetypeofinteractionandtheobserver’schoiceabouthowtoclassifytheobservation.Forexample,passive/active(twocategories);introvert/extrovert(twocategories);always/sometimes/never(threecategories);stronglyagree/agree/uncertain/disagree/stronglydisagree(fivecategories).Theuseofcategoriestorecordanobservationmaysufferfromthesameproblemsasthoseassociatedwithscales.Recordingonelectronicdevices–Observationcanalsoberecordedonvideotapeorotherelectronicdevicesandthenanalysed.Theadvantageofrecordinganinteractioninthiswayisthattheobservercanseeitanumberoftimesbeforeinterpretinganinteractionordrawinganyconclusionsfromitandcanalsoinviteotherprofessionalstoviewtheinteractioninordertoarriveatmoreobjectiveconclusions.However,oneofthedisadvantagesisthatsomepeoplemayfeeluncomfortableormaybehavedifferentlybeforeacamera.Thereforetheinteractionmaynotbeatruereflectionofthesituation.
FIGURE9.2Athree-directionalratingscale
Thechoiceofaparticularmethodforrecordingyourobservationisdependentuponthepurposeofthe observation, the complexity of the interaction and the type of population being observed. It isimportanttoconsiderthesefactorsbeforedecidinguponthemethodforrecordingyourobservation.
Theinterview
Interviewingisacommonlyusedmethodofcollectinginformationfrompeople.Inmanywalksoflifewecollectinformationthroughdifferentformsofinteractionwithothers.Therearemanydefinitionsofinterviews. According to Monette et al. (1986: 156), ‘an interview involves an interviewer readingquestionstorespondentsandrecordingtheiranswers’.AccordingtoBurns(1997:329),‘aninterviewisaverbalinterchange,oftenfacetoface,thoughthetelephonemaybeused,inwhichaninterviewertriestoelicitinformation,beliefsoropinionsfromanotherperson’.Anyperson-to-personinteraction,eitherfacetofaceorotherwise,betweentwoormoreindividualswithaspecificpurposeinmindiscalledaninterview.When interviewing a respondent, you, as a researcher, have the freedom to decide the format and
contentofquestionstobeaskedofyourrespondents,selectthewordingofyourquestions,decidethewayyouwanttoaskthemandchoosetheorderinwhichtheyaretobeasked.Thisprocessofaskingquestionscanbeeitherveryflexible,whereyouastheinterviewerhavethefreedomtothinkaboutandformulatequestionsastheycometoyourmindaroundtheissuebeinginvestigated,orinflexible,whereyouhavetokeepstrictlytothequestionsdecidedbeforehand–includingtheirwording,sequenceandthemannerinwhichtheyareasked.InterviewsareclassifiedintodifferentcategoriesaccordingtothisdegreeofflexibilityasinFigure9.3.
FIGURE9.3Typesofinterview
UnstructuredInterviews
Thestrengthofunstructuredinterviewsisthealmostcompletefreedomtheyprovideintermsofcontentand structure. You are free to order these in whatever sequence you wish. You also have completefreedomintermsofthewordingyouuseandthewayyouexplainquestionstoyourrespondents.Youmayformulatequestionsandraise issueson thespurof themoment,dependinguponwhatoccurs to
youinthecontextofthediscussion.Unstructuredinterviewsareprevalentinbothquantitativeandqualitativeresearch.Thedifferenceis
in how information obtained through them in response to your questions is likely to be used. Inquantitative researchyoudevelop response categorisations from responseswhich are then coded andquantified.Inqualitativeresearchtheresponsesareusedasdescriptors,ofteninverbatimform,andcanbeintegratedwithyourarguments, flowofwritingandsequenceof logic.Asunstructuredinterviewsaredominantlyusedinqualitativeresearch,theyaredescribedingreaterdetailunder‘Methodsofdatacollectioninqualitativeresearch’laterinthischapter.
Structuredinterviews
Inastructuredinterviewtheresearcherasksapredeterminedsetofquestions,usingthesamewordingandorderofquestionsasspecifiedintheinterviewschedule.Aninterviewscheduleisawrittenlistofquestions,openendedorclosed,prepared forusebyan interviewer inaperson-to-person interaction(thismaybefacetoface,bytelephoneorbyotherelectronicmedia).Notethataninterviewscheduleisaresearchtool/instrumentforcollectingdata,whereasinterviewingisamethodofdatacollection.Oneofthemainadvantagesofthestructuredinterviewisthatitprovidesuniforminformation,which
assuresthecomparabilityofdata.Structuredinterviewingrequiresfewerinterviewingskillsthandoesunstructuredinterviewing.
Thequestionnaire
Aquestionnaire isawrittenlistofquestions,theanswerstowhicharerecordedbyrespondents.Inaquestionnaire respondents read the questions, interpret what is expected and then write down theanswers.Theonlydifferencebetweenaninterviewscheduleandaquestionnaireisthatintheformeritistheinterviewerwhoasksthequestions(andifnecessary,explainsthem)andrecordstherespondent’srepliesonaninterviewschedule,andinthelatterrepliesarerecordedbytherespondents themselves.This distinction is important in accounting for the respective strengths and weaknesses of the twomethods.Inthecaseofaquestionnaire,asthereisnoonetoexplainthemeaningofquestionstorespondents,it
is important that the questions are clear and easy to understand.Also, the layout of a questionnaireshouldbesuchthatitiseasytoreadandpleasanttotheeye,andthesequenceofquestionsshouldbeeasy to follow.A questionnaire should be developed in an interactive style. Thismeans respondentsshouldfeelasifsomeoneistalkingtothem.Inaquestionnaire,asensitivequestionoraquestionthatrespondents may feel hesitant about answering should be prefaced by an interactive statementexplainingtherelevanceofthequestion.Itisagoodideatouseadifferentfontforthesestatementstodistinguish themfrom theactualquestions.Examples inFigures9.4 and9.5 taken from two surveysrecentlycarriedoutbytheauthorwiththehelpoftwostudentsexplainsomeoftheabovepoints.
FIGURE9.4Example1
Waysofadministeringaquestionnaire
Aquestionnairecanbeadministeredindifferentways.
FIGURE9.5Example2
Themailedquestionnaire–Themostcommonapproachtocollectinginformationistosendthequestionnairetoprospectiverespondentsbymail.Obviouslythisapproachpresupposesthatyouhaveaccesstotheiraddresses.Usuallyitisagoodideatosendaprepaid,self-addressedenvelopewiththequestionnaireasthismightincreasetheresponserate.Amailedquestionnairemustbeaccompaniedbyacoveringletter(seebelowfordetails).Oneofthemajorproblemswiththismethodisthelowresponserate.Inthecaseofanextremelylowresponserate,thefindingshaveverylimitedapplicabilitytothepopulationstudied.Collectiveadministration–Oneofthebestwaysofadministeringaquestionnaireistoobtainacaptiveaudiencesuchasstudentsinaclassroom,peopleattendingafunction,participantsinaprogrammeorpeopleassembledinoneplace.Thisensuresaveryhighresponserateasyouwillfindfewpeoplerefusetoparticipateinyourstudy.Also,asyouhavepersonalcontactwiththestudypopulation,youcanexplainthepurpose,relevanceandimportanceofthestudyandcanclarifyanyquestionsthatrespondentsmayhave.Theauthor’sadviceisthatifyouhaveacaptiveaudienceforyourstudy,don’tmisstheopportunity–itisthequickestwayofcollectingdata,ensuresaveryhighresponserateandsavesyoumoneyonpostage.Administrationinapublicplace–Sometimesyoucanadministeraquestionnaireinapublicplacesuchasashoppingcentre,healthcentre,hospital,schoolorpub.Ofcoursethisdependsuponthetypeofstudypopulationyouarelookingforandwhereitislikelytobefound.Usuallythepurposeofthestudyisexplainedtopotentialrespondentsastheyapproachandtheirparticipationinthestudyisrequested.Apartfrombeingslightlymoretimeconsuming,thismethodhasalltheadvantagesofadministeringaquestionnairecollectively.
Choosingbetweenaninterviewandaquestionnaire
Thechoicebetweenaquestionnaireandan interviewschedule is importantandshouldbeconsideredthoroughlyasthestrengthsandweaknessesofthetwomethodscanaffectthevalidityofthefindings.The nature of the investigation and the socioeconomic–demographic characteristics of the studypopulationarecentralinthischoice.Theselectionbetweenaninterviewscheduleandaquestionnaireshouldbebaseduponthefollowingcriteria:
Thenatureoftheinvestigation–Ifthestudyisaboutissuesthatrespondentsmayfeelreluctanttodiscusswithaninvestigator,aquestionnairemaybethebetterchoiceasitensuresanonymity.Thismaybethecasewithstudiesondruguse,sexuality,indulgenceincriminalactivitiesandpersonalfinances.However,therearesituationswherebetterinformationaboutsensitiveissuescanbeobtainedbyinterviewingrespondents.Itdependsonthetypeofstudypopulationandtheskillsoftheinterviewer.Thegeographicaldistributionofthestudypopulation–Ifpotentialrespondentsarescatteredoverawidegeographicalarea,youhavenochoicebuttouseaquestionnaire,asinterviewinginthesecircumstanceswouldbeextremelyexpensive.Thetypeofstudypopulation–Ifthestudypopulationisilliterate,veryyoungorveryold,orhandicapped,theremaybenooptionbuttointerviewrespondents.
Advantagesofaquestionnaire
Aquestionnairehasseveraladvantages:
Itislessexpensive.Asyoudonotinterviewrespondents,yousavetime,andhumanandfinancialresources.Theuseofaquestionnaire,therefore,iscomparativelyconvenientandinexpensive.Particularlywhenitisadministeredcollectivelytoastudypopulation,itisanextremelyinexpensivemethodofdatacollection.Itoffersgreateranonymity.Asthereisnoface-to-faceinteractionbetweenrespondentsandinterviewer,thismethodprovidesgreateranonymity.Insomesituationswheresensitivequestionsareaskedithelpstoincreasethelikelihoodofobtainingaccurateinformation.
Disadvantagesofaquestionnaire
Althoughaquestionnairehasseveraldisadvantages, it is important tonote thatnotalldatacollectionusingthismethodhasthesedisadvantages.Theprevalenceofadisadvantagedependsonanumberoffactors,butyouneedtobeawareofthemtounderstandtheirpossiblebearingonthequalityofthedata.Theseare:
Applicationislimited.Onemaindisadvantageisthatapplicationislimitedtoastudypopulationthatcanreadandwrite.Itcannotbeusedonapopulationthatisilliterate,veryyoung,veryoldorhandicapped.Responserateislow.Questionnairesarenotoriousfortheirlowresponserates;thatis,peoplefailtoreturnthem.Ifyouplantouseaquestionnaire,keepinmindthatbecausenoteveryonewillreturntheirquestionnaire,yoursamplesizewillineffectbereduced.Theresponseratedependsuponanumberoffactors:theinterestofthesampleinthetopicofthestudy;thelayoutandlengthofthequestionnaire;thequalityoftheletterexplainingthepurposeandrelevanceofthestudy;andthemethodologyusedtodeliverthequestionnaire.Youshouldconsideryourselfluckytoobtaina50percentresponserateandsometimesitmaybeaslowas20percent.However,asmentioned,theresponserateisnotaproblemwhenaquestionnaireisadministeredinacollectivesituation.Thereisaself-selectingbias.Noteveryonewhoreceivesaquestionnairereturnsit,sothereisaself-selectingbias.Thosewhoreturntheirquestionnairemayhaveattitudes,attributesormotivationsthataredifferentfromthosewhodonot.Hence,iftheresponserateisverylow,thefindingsmaynotberepresentativeofthetotalstudypopulation.Opportunitytoclarifyissuesislacking.If,foranyreason,respondentsdonotunderstandsomequestions,thereisalmostnoopportunityforthemtohavethemeaningclarifiedunlesstheygetintouchwithyou–theresearcher(whichdoesnothappenoften).Ifdifferentrespondentsinterpretquestionsdifferently,thiswillaffectthequalityoftheinformationprovided.Spontaneousresponsesarenotallowedfor.Mailedquestionnairesareinappropriatewhenspontaneousresponsesarerequired,asaquestionnairegivesrespondentstimetoreflectbeforeanswering.Theresponsetoaquestionmaybeinfluencedbytheresponsetootherquestions.Asrespondentscanreadallthequestionsbeforeanswering(whichusuallyhappens),thewaytheyansweraparticularquestionmaybeaffectedbytheirknowledgeofotherquestions.Itispossibletoconsultothers.Withmailedquestionnairesrespondentsmayconsultotherpeople
beforeresponding.Insituationswhereaninvestigatorwantstofindoutonlythestudypopulation’sopinions,thismethodmaybeinappropriate,thoughrequestingrespondentstoexpresstheirownopinionmayhelp.Aresponsecannotbesupplementedwithotherinformation.Aninterviewcansometimesbesupplementedwithinformationfromothermethodsofdatacollectionsuchasobservation.However,aquestionnairelacksthisadvantage.
Advantagesoftheinterview
Theinterviewismoreappropriateforcomplexsituations.Itisthemostappropriateapproachforstudyingcomplexandsensitiveareasastheinterviewerhastheopportunitytopreparearespondentbeforeaskingsensitivequestionsandtoexplaincomplexonestorespondentsinperson.Itisusefulforcollectingin-depthinformation.Inaninterviewsituationitispossibleforaninvestigatortoobtainin-depthinformationbyprobing.Hence,insituationswherein-depthinformationisrequired,interviewingisthepreferredmethodofdatacollection.Informationcanbesupplemented.Aninterviewerisabletosupplementinformationobtainedfromresponseswiththosegainedfromobservationofnon-verbalreactions.Questionscanbeexplained.Itislesslikelythataquestionwillbemisunderstoodastheinterviewercaneitherrepeataquestionorputitinaformthatisunderstoodbytherespondent.Interviewinghasawiderapplication.Aninterviewcanbeusedwithalmostanytypeofpopulation:children,thehandicapped,illiterateorveryold.
Disadvantagesoftheinterview
Interviewingistimeconsumingandexpensive.Thisisespeciallysowhenpotentialrespondentsarescatteredoverawidegeographicalarea.However,ifyouhaveasituationsuchasanoffice,ahospitaloranagencywherepotentialrespondentscometoobtainaservice,interviewingtheminthatsettingmaybelessexpensiveandlesstimeconsuming.Thequalityofdatadependsuponthequalityoftheinteraction.Inaninterviewthequalityofinteractionbetweenaninterviewerandintervieweeislikelytoaffectthequalityoftheinformationobtained.Also,becausetheinteractionineachinterviewisunique,thequalityoftheresponsesobtainedfromdifferentinterviewsmayvarysignificantly.Thequalityofdatadependsuponthequalityoftheinterviewer.Inaninterviewsituationthequalityofthedatageneratedisaffectedbytheexperience,skillsandcommitmentoftheinterviewer.Thequalityofdatamayvarywhenmanyinterviewersareused.Useofmultipleinterviewersmaymagnifytheproblemsidentifiedinthetwopreviouspoints.Theresearchermayintroducehis/herbias.Researcherbiasintheframingofquestionsandtheinterpretationofresponsesisalwayspossible.Iftheinterviewsareconductedbyapersonorpersons,paidorvoluntary,otherthantheresearcher,itisalsopossiblethattheymayexhibitbiasinthewaytheyinterpretresponses,selectresponsecategoriesorchoosewordstosummariserespondents’expressedopinions.
Contentsofthecoveringletter
Itisessentialthatyouwriteacoveringletterwithyourmailedquestionnaire.Itshouldverybriefly:
introduceyouandtheinstitutionyouarerepresenting;describeintwoorthreesentencesthemainobjectivesofthestudy;explaintherelevanceofthestudy;conveyanygeneralinstructions;indicatethatparticipationinthestudyisvoluntary–ifrecipientsdonotwanttorespondtothequestionnaire,theyhavetherightnotto;assurerespondentsoftheanonymityoftheinformationprovidedbythem;provideacontactnumberincasetheyhaveanyquestions;giveareturnaddressforthequestionnaireandadeadlineforitsreturn;thankthemfortheirparticipationinthestudy.
Formsofquestion
Theformandwordingofquestionsusedinanintervieworaquestionnaireareextremelyimportantinaresearch instrument as they have an effect on the type and quality of information obtained from arespondent.Thewordingandstructureofquestionsshouldthereforebeappropriate,relevantandfreefromanyof theproblemsdiscussedinthesectiontitled‘Formulatingeffectivequestions’ later inthischapter.Beforethis,letusdiscussthetwoformsofquestions,openendedandclosed,whicharebothcommonlyusedinsocialsciencesresearch.Inanopen-endedquestionthepossibleresponsesarenotgiven.Inthecaseofaquestionnaire,the
respondent writes down the answers in his/her words, but in the case of an interview schedule theinvestigator records the answers either verbatim or in a summary. In a closed question the possibleanswers are set out in the questionnaire or schedule and the respondent or the investigator ticks thecategory that best describes the respondent’s answer. It is usually wise to provide a category‘Other/please explain’ to accommodate any response not listed. The questions in Figure 9.6 areclassifiedasclosedquestions.Thesamequestionscouldbeaskedasopen-endedquestions,asshowninFigure9.7.Whendecidingwhethertouseopen-endedorclosedquestionstoobtaininformationaboutavariable,
visualisehowyouplantousetheinformationgenerated.Thisisimportantbecausethewayyouframeyourquestionsdeterminestheunitofmeasurementwhichcouldbeusedtoclassifytheresponses.Theunitofmeasurementinturndictateswhatstatisticalprocedurescanbeappliedtothedataandthewaytheinformationcanbeanalysedanddisplayed.Letustake,asanexample,thequestionaboutthevariable:‘income’.Inclosedquestionsincomecan
bequalitativelyrecordedincategoriessuchas‘aboveaverage/average/belowaverage’,orquantitativelyincategoriessuchas‘under$10000/$10000–$19999/…’.Yourchoiceofqualitativeandquantitativecategoriesaffectstheunitofmeasurementforincome(qualitativeusestheordinalscaleandquantitativetheratioscaleofmeasurement),whichinturnwillaffect theapplicationofstatisticalprocedures.Forexample,youcannotcalculate theaverageincomeofapersonfromtheresponsestoquestionC(a) inFigure9.6;norcanyoucalculatethemedianormodalcategoryofincome.Butfromtheresponsestoquestion C, you can accurately calculate modal category of income. However, the average and themedian income cannot be accurately calculated (such calculations are usually made under certain
assumptions).From the responses to questionC inFigure9.7,where the income for a respondent isrecorded in exact dollars, the different descriptors of income can be calculated very accurately. Inaddition,informationonincomecanbedisplayedinanyform.Youcancalculatetheaverage,medianormode. The same is true for any other information obtained in response to open-ended and closedquestions.
FIGURE9.6ExamplesofclosedquestionsInclosedquestions,havingdevelopedcategories,youcannotchangethem;hence,youshouldbevery
certain about your categories when developing them. If you ask an open-ended question, you candevelopanynumberofcategoriesatthetimeofanalysis.Both open-ended and closed questions have their advantages and disadvantages in different
situations. To some extent, their advantages and disadvantages depend uponwhether they are beingusedinanintervieworinaquestionnaireandonwhethertheyarebeingusedtoseekinformationaboutfactsoropinions.Asarule,closedquestionsareextremelyusefulforelicitingfactualinformationandopen-ended questions for seeking opinions, attitudes and perceptions. The choice of open-ended orclosed questions should bemade according to the purpose forwhich a piece of information is to beused,thetypeofstudypopulationfromwhichinformationisgoingtobeobtained,theproposedformatforcommunicatingthefindingsandthesocioeconomicbackgroundofthereadership.
FIGURE9.7Examplesofopen-endedquestions
Advantagesanddisadvantagesofopen-endedquestions
Open-endedquestionsprovidein-depthinformationifusedinaninterviewbyanexperiencedinterviewer.Inaquestionnaire,open-endedquestionscanprovideawealthofinformationprovidedrespondentsfeelcomfortableaboutexpressingtheiropinionsandarefluentinthelanguageused.Ontheotherhand,analysisofopen-endedquestionsismoredifficult.Theresearcherusuallyneedstogothroughanotherprocess–contentanalysis–inordertoclassifythedata.Inaquestionnaire,open-endedquestionsproviderespondentswiththeopportunitytoexpressthemselvesfreely,resultinginagreatervarietyofinformation.Thusrespondentsarenot‘conditioned’byhavingtoselectanswersfromalist.Thedisadvantageoffreechoiceisthat,inaquestionnaire,somerespondentsmaynotbeabletoexpressthemselves,andsoinformationcanbelost.Asopen-endedquestionsallowrespondentstoexpressthemselvesfreely,theyvirtuallyeliminatethepossibilityofinvestigatorbias(investigatorbiasisintroducedthroughtheresponsepatternpresentedtorespondents).Ontheotherhand,thereisagreaterchanceofinterviewerbiasinopen-endedquestions.
Advantagesanddisadvantagesofclosedquestions
Oneofthemaindisadvantagesofclosedquestionsisthattheinformationobtainedthroughthemlacksdepthandvariety.Thereisagreaterpossibilityofinvestigatorbiasbecausetheresearchermaylistonlytheresponsepatternsthats/heisinterestedinorthosethatcometomind.Evenifthecategoryof‘other’isoffered,mostpeoplewillusuallyselectfromthegivenresponses,andsothefindingsmaystillreflectresearcherbias.Inaquestionnaire,thegivenresponsepatternforaquestioncouldconditionthethinkingofrespondents,andsotheanswersprovidedmaynottrulyreflectrespondents’opinions.Rather,theymayreflecttheextentofagreementordisagreementwiththeresearcher’sopinionoranalysisofasituation.Theeaseofansweringaready-madelistofresponsesmaycreateatendencyamongsomerespondentsandinterviewerstotickacategoryorcategorieswithoutthinkingthroughtheissue.Closedquestions,becausetheyprovide‘ready-made’categorieswithinwhichrespondentsreplytothequestionsaskedbytheresearcher,helptoensurethattheinformationneededbytheresearcherisobtainedandtheresponsesarealsoeasiertoanalyse.
Formulatingeffectivequestions
Thewordingandtoneofyourquestionsare importantbecausethe informationanditsquality largelydependuponthesefactors.Itisthereforeimportanttobecarefulaboutthewayyouformulatequestions.Thefollowingaresomeconsiderationstokeepinmindwhenformulatingquestions:
•Alwaysusesimpleandeverydaylanguage.Yourrespondentsmaynotbehighlyeducated,andeveniftheyaretheystillmaynotknowsomeofthe‘simple’technicaljargonthatyouareusedto.Particularlyinaquestionnaire,takeextracaretousewordsthatyourrespondentswillunderstandasyouwillhaveno opportunity to explain questions to them. A pre-test should show you what is and what is notunderstoodbyyourrespondents.Forexample:
Isanyoneinyourfamilyadipsomaniac?(Bailey1978:100)
In this question many respondents, even some who are well educated, will not understand‘dipsomaniac’and,hence,theyeitherdonotansweroranswerthequestionwithoutunderstanding.
• Do not use ambiguous questions. An ambiguous question is one that contains more than onemeaning and that can be interpreted differently by different respondents.Thiswill result in differentanswers,makingitdifficult,ifnotimpossible,todrawanyvalidconclusionsfromtheinformation.Thefollowingquestionshighlighttheproblem:
Is your work made difficult because you are expecting a baby? (Moser & Kalton 1989:323)Yes No
Inthesurveyallwomenwereaskedthisquestion.Thosewomenwhowerenotpregnantticked‘No’,meaning no they were not pregnant, and those who were pregnant and who ticked ‘No’ meantpregnancyhadnotmadetheirworkdifficult.Thequestionhasotherambiguitiesaswell:itdoesnotspecifythetypeofworkandthestageofpregnancy.
Areyousatisfiedwithyourcanteen?(Moser&Kalton1989:319)
Thisquestionisalsoambiguousasitdoesnotaskrespondentstoindicatetheaspectsofthecanteenwith which they may be satisfied or dissatisfied. Is it with the service, the prices, the physicalfacilities,theattitudeofthestafforthequalityofthemeals?Respondentsmayhaveanyoneoftheseaspectsinmindwhentheyanswerthequestion.Orthequestionshouldhavebeenwordeddifferentlylike,‘Areyou,onthewhole,satisfiedwithyourcanteen?’
•Donotaskdouble-barrelledquestions.Adouble-barrelledquestionisaquestionwithinaquestion.Themain problemwith this type of question is that one does not knowwhich particular question arespondent has answered. Some respondentsmay answer both parts of the question and othersmayansweronlyoneofthem.
Howoftenandhowmuchtimedoyouspendoneachvisit?
This questionwas asked in a survey inWesternAustralia to ascertain the need for child-mindingservicesinoneofthehospitals.Thequestionhastwoparts:howoftendoyouvisitandhowmuchtime is spent on each visit? In this type of question some respondentsmay answer the first part,
whereas others may answer the second part and some may answer both parts. Incidentally, thisquestionisalsoambiguousinthatitdoesnotspecify‘howoften’intermsofaperiodoftime.Isitinaweek,afortnight,amonthorayear?
Doesyourdepartmenthaveaspecialrecruitmentpolicyforracialminoritiesandwomen?(Bailey1978:97)
This question is double barrelled in that it asks respondents to indicatewhether their office has aspecialrecruitmentpolicyfortwopopulationgroups:racialminoritiesandwomen.A‘yes’responsedoesnotnecessarilymeanthattheofficehasaspecialrecruitmentpolicyforbothgroups.
•Donotaskleadingquestions.Aleadingquestionisonewhich,byitscontents,structureorwording,leads a respondent to answer in a certain direction. Such questions are judgemental and leadrespondentstoanswereitherpositivelyornegatively.
Unemploymentisincreasing,isn’tit?
Smokingisbad,isn’tit?
Thefirstproblemisthatthesearenotquestionsbutstatements.Becausethestatementssuggestthat‘unemploymentisincreasing’and‘smokingisbad’,respondentsmayfeelthattodisagreewiththemis to be in thewrong, especially if they feel that the researcher is an authority and that if s/he issayingthat‘unemploymentisincreasing’or‘smokingisbad’,itmustbeso.Thefeelingthatthereisa‘right’answercan‘force’peopletorespondinawaythatiscontrarytotheirtrueposition.
•Donotaskquestionsthatarebasedonpresumptions.Insuchquestionstheresearcherassumesthatrespondentsfitintoaparticularcategoryandseeksinformationbaseduponthatassumption.
Howmanycigarettesdoyousmokeinaday?(Moser&Kalton1989:325)
Whatcontraceptivesdoyouuse?
Both thesequestionswereaskedwithoutascertainingwhetherornotrespondentsweresmokersorsexuallyactive.Insituationslikethisitisimportanttoascertainfirstwhetherornotarespondentfitsintothecategoryaboutwhichyouareenquiring.
Constructingaresearchinstrumentinquantitativeresearch
Theconstructionofaresearchinstrumentortoolisanextremelyimportantaspectofaresearchprojectbecauseanythingyousaybywayoffindingsorconclusionsisbaseduponthetypeofinformationyoucollect, and the data you collect is entirely dependent upon the questions that you ask of yourrespondents.Thefamoussayingaboutcomputers–‘garbagein,garbageout’–isalsoapplicabletodatacollection.Theresearchtoolprovidestheinputtoastudyandthereforethequalityandvalidityoftheoutput,thefindings,aresolelydependentuponit.Inspiteofitsimmenseimportance,totheauthor’sknowledge,nospecificguidelinesforbeginnerson
howtoconstructaresearchtoolexist.Studentsare left to learnfor themselvesunder theguidanceoftheir research supervisor. The guidelines suggested below outline a broad approach, especially forbeginners.The underlying principle is to ensure the validity of your instrument bymaking sure thatyour questions relate to the objectives of your study. Therefore, clearly defined objectives play an
extremely important role as each question in the instrumentmust stem from the objectives, researchquestions and/or hypotheses of the study. It is suggested that a beginner should adopt the followingprocedure:
StepI Ifyouhavenotalreadydoneso,clearlydefineandindividuallylistallthespecificobjectives,researchquestionsorhypotheses,ifany,tobetested.
StepII Foreachobjective,researchquestionorhypothesis,listalltheassociatedquestionsthatyouwanttoanswerthroughyourstudy.StepIII TakeeachquestionthatyouidentifiedinStepIIandlisttheinformationrequiredtoanswerit.
StepIV
Formulatequestion(s)thatyouwanttoaskofyourrespondentstoobtaintherequiredinformation.
In the above process you may find that the same piece of information is required for a number ofquestions.Insuchasituationthequestionshouldbeaskedonceonly.Tounderstandthisprocess,seeTable9.1forwhichwehavealreadydevelopedasetofobjectivesinFigure4.4inChapter4.
Askingpersonalandsensitivequestions
In the social sciences, sometimes one needs to ask questions that are of a personal nature. Somerespondentsmayfind thisoffensive. It is important tobeawareof thisas itmayaffect thequalityofinformation or even result in an interview being terminated or questionnaires not being returned.Researchershaveusedanumberofapproachestodealwiththisproblembutitisdifficulttosaywhichapproachisbest.AccordingtoBradburnandSudman:
nodatacollectionmethodissuperiortoothermethodsforalltypesofthreateningquestions.Ifoneaccepts the results at face value, each of the data gathering methods is best under certainconditions.(1979:12–13)
TABLE9.1Guidelinesforconstructingaresearchinstrument(quantitativeresearch):astudytoevaluatecommunityresponsivenessinahealthprogramme
Intermsofthebesttechniqueforaskingsensitiveorthreateningquestions,thereappearstobetwooppositeopinions,basedonthemannerinwhichthequestionisasked:
1. adirectmanner;2. anindirectmanner.
The advantagewith the first approach is that one canbe sure that an affirmative answer is accurate.Thosewhoadvocatethesecondapproachbelievethatdirectquestioningislikelytooffendrespondentsandhencetheyareunlikelytoanswereventhenon-sensitivequestions.Somewaysofaskingpersonalquestionsinanindirectmannerareasfollows:
byshowingdrawingsorcartoons;byaskingarespondenttocompleteasentence;byaskingarespondenttosortcardscontainingstatements;byusingrandomdevices.
Todescribethesemethodsindetailisbeyondthescopeofthisbook.
Theorderofquestions
Theorderofquestionsinaquestionnaireorinaninterviewscheduleisimportantasitaffectsthequalityof information,and the interestandevenwillingnessofarespondent toparticipate inastudy.Again,there are two categories of opinion as to the bestway to order questions. The first is that questionsshouldbeaskedinarandomorderandthesecondisthattheyshouldfollowalogicalprogressionbasedupontheobjectivesof thestudy.Theauthorbelieves that the latterprocedure isbetteras itgraduallyleadsrespondentsintothethemesofthestudy,startingwithsimplethemesandprogressingtocomplexones.This approach sustains the interest of respondents andgradually stimulates them to answer thequestions.However,therandomapproachisusefulinsituationswherearesearcherwantsrespondentsto express their agreement or disagreement with different aspects of an issue. In this case a logicallisting of statements or questions may ‘condition’ a respondent to the opinions expressed by theresearcherthroughthestatements.
Pre-testingaresearchinstrument
Having constructed your research instrument,whether an interview schedule or a questionnaire, it isimportantthatyoutestitoutbeforeusingitforactualdatacollection.Pre-testingaresearchinstrumententailsacriticalexaminationoftheunderstandingofeachquestionanditsmeaningasunderstoodbyarespondent.Apre-testshouldbecarriedoutunderactualfieldconditionsonagroupofpeoplesimilartoyour study population. The purpose is not to collect data but to identify problems that the potentialrespondentsmight have in either understanding or interpreting a question.Your aim is to identify ifthere are problems in understanding theway a questionhas beenworded, the appropriateness of themeaningitcommunicates,whetherdifferentrespondentsinterpretaquestiondifferently,andtoestablishwhether their interpretation isdifferent towhatyouwere trying toconvey. If thereareproblemsyouneedtore-examinethewordingtomakeitclearerandunambiguous.
Prerequisitesfordatacollection
Beforeyoustartobtaininginformationfrompotentialrespondentsitisimperativethatyoumakesureoftheir:
motivationtosharetherequiredinformation–Itisessentialforrespondentstobewillingtoshareinformationwithyou.Youshouldmakeeveryefforttomotivatethembyexplainingclearlyandinsimpletermstheobjectivesandrelevanceofthestudy,eitheratthetimeoftheintervieworinthecoveringletteraccompanyingthequestionnaireand/orthroughinteractivestatementsinthequestionnaire.clearunderstandingofthequestions–Respondentsmustunderstandwhatisexpectedoftheminthequestions.Ifrespondentsdonotunderstandaquestionclearly,theresponsegivenmaybeeitherwrongorirrelevant,ormakenosense.possessionoftherequiredinformation–Thethirdprerequisiteisthatrespondentsmusthavetheinformationsought.Thisisofparticularimportancewhenyouareseekingfactualortechnicalinformation.Ifrespondentsdonothavetherequiredinformation,theycannotprovideit.
Methodsofdatacollectioninqualitativeresearch
To draw a clear distinction between quantitative and qualitative methods of data collection is bothdifficultandinappropriatebecauseoftheoverlapbetweenthem.Thedifferencebetweenthemmainlylies in themanner inwhich amethod is applied in an actual data collection situation.Use of thesemethodsinquantitativeresearchdemandsstandardisationofquestionstobeaskedoftherespondents,arigidadherencetotheirstructureandorder,anadoptionofaprocessthatistestedandpredetermined,andmaking sure of the validity and reliability of the process aswell as the questions.However, themethods of data collection in qualitative research follow a convention which is almost opposite toquantitativeresearch.Thewording,orderandformatofthesequestionsareneitherpredeterminednorstandardised.Qualitativemethodsarecharacterisedbyflexibilityandfreedomintermsofstructureandordergiventotheresearcher.Asmentionedinthepreviouschapter,mostqualitativestudydesignsaremethodbased: that is, the
method of data collection seems to determine the design. In some situations it becomes difficult toseparate a study design from the method of data collection. For example, in-depth interviewing,narrativesandoralhistoryarebothdesignsandmethodsofdatacollection.Thismayconfusesomebutheretheyaredetailedasmethodsandnotdesigns.Therearethreemainmethodsofdatacollectioninqualitativeresearch:
1. unstructuredinterviews;2. participantobservation;3. secondarysources.
Participantobservationhasbeenadequatelycoveredearlierinthischapterandsecondarysourceswillbecoveredinalatersection,soatthispointwewillfocusonunstructuredinterviews,whicharebyfarthemostcommonlyusedmethodofdatacollectioninqualitativeresearch.Flexibility,freedomandspontaneityincontentsandstructureunderpinaninteractioninalltypesof
unstructuredinterview.Thisinteractioncanbeataone-to-one(researcherandarespondent)oragroup(researcherandagroupofrespondents)level.Thereareseveraltypesofunstructuredinterviewthatareprevalent in qualitative research, for example in-depth interviewing, focus group interviewing,narrativesandoralhistories.Belowisabriefdescriptionofeachofthem.ForadetailedunderstandingreadersshouldconsulttherelevantreferenceslistedintheBibliography.
In-depthinterviews
The theoretical roots of in-depth interviewing are in what is known as the interpretive tradition.According toTaylor andBogdan, in-depth interviewing is ‘repeated face-to-face encounters betweenthe researcher and informants directed towards understanding informants’ perspectives on their lives,experiences,orsituationsasexpressedin theirownwords’(1998:77).Thisdefinitionunderlines twoessential characteristics of in-depth interviewing: (1) it involves face-to-face, repeated interactionbetweentheresearcherandhis/herinformant(s);and(2)itseekstounderstandthelatter’sperspectives.Because thismethod involves repeated contacts and hence an extended length of time spentwith aninformant, it isassumedthattherapportbetweenresearcherandinformantwillbeenhanced,andthatthe corresponding understanding and confidence between the twowill lead to in-depth and accurateinformation.
Focusgroupinterviews
Theonly difference between a focus group interview and an in-depth interview is that the former isundertakenwithagroupandthelatterwithanindividual.Inafocusgroupinterview,youexploretheperceptions, experiences and understandings of a group of people who have some experience incommonwithregardtoasituationorevent.Forexample,youmayexplorewithrelevantgroupssuchissuesasdomesticviolence,physicaldisabilityorrefugees.Infocusgroupinterviews,broaddiscussiontopicsaredevelopedbeforehand,eitherbytheresearcher
or by the group. These provide a broad frame for discussionswhich follow.The specific discussionpoints emerge as a part of the discussion. Members of a focus group express their opinions whilediscussingtheseissues.You,asaresearcher,needtoensure thatwhatever isexpressedordiscussedisrecordedaccurately.
Usethemethodofrecordingthatsuitsyouthebest.Youmayaudiotapediscussions,employsomeoneelsetorecordthemorrecordthemyourselfimmediatelyaftereachsession.Ifyouaretakingyourownnotesduringdiscussions,youneedtobecarefulnottolosesomethingofimportancebecauseofyourinvolvementindiscussions.Youcanandshouldtakeyourwrite-upondiscussionsbacktoyourfocusgroupforcorrection,verificationandconfirmation.
Narratives
The narrative technique of gathering information has even less structure than the focus group.Narrativeshavealmostnopredeterminedcontentsexceptthattheresearcherseekstohearaperson’sretellingofanincidentorhappeninginhis/herlife.Essentially,thepersontellshis/herstoryaboutanincident or situation and you, as the researcher, listen passively. Occasionally, you encourage theindividual by using active listening techniques; that is, you say words such as ‘uh huh’, ‘mmmm’,‘yeah’,‘right’andnodasappropriate.Basically,youletthepersontalkfreelyandwithoutinterrupting.Narrativesareaverypowerfulmethodofdatacollectionforsituationswhicharesensitiveinnature.
Forexample,youmaywanttofindoutabouttheimpactofchildsexualabuseonpeoplewhohavegonethroughsuchanexperience.You,asaresearcher,askthesepeopletonarratetheirexperiencesandhowtheyhavebeen affected.Narrativesmayhave a therapeutic impact; that is, sometimes simply tellingtheirstorymayhelpapersontofeelmoreateasewiththeevent.Sometherapistsspecialiseinnarrativetherapy.Buthere,weareconcernedwithnarrativesasamethodofdatacollection.Aswith focus group interviews, you need to choose the recording system that suits you the best.
Havingcompletednarrativesessionsyouneedtowriteyourdetailednotesandgivethembacktotherespondenttocheckforaccuracy.
Oralhistories
Oral histories, like narratives, involve the use of both passive and active listening. Oral histories,however,aremorecommonlyusedforlearningaboutahistoricaleventorepisodethattookplaceinthepastorforgaininginformationaboutacultural,customorstorythathasbeenpassedfromgenerationtogeneration. Narratives are more about a person’s personal experiences whereas historical, social orculturaleventsarethesubjectsoforalhistories.Supposeyouwant to findoutabout the lifeafter theSecondWorldWar insomeregional townof
WesternAustraliaoraboutthelivingconditionsofAboriginalandTorresStraitIslanderpeopleinthe1960s.Youwouldtalktopersonswhowerealiveduringthatperiodandaskthemaboutlifeatthattime.Datacollection throughunstructured interviewing isextremelyuseful insituationswhereeither in-
depthinformationisneededorlittleisknownaboutthearea.Theflexibilityallowedtotheinterviewerinwhats/heasksofarespondentisanassetasitcanelicitextremelyrichinformation.Asitprovidesin-depth information, this technique is used bymany researchers for constructing a structured researchinstrument.On the other hand, since an unstructured interview does not list specific questions to beasked of respondents, the comparability of questions asked and responses obtained may become aproblem.Astheresearchergainsexperienceduringtheinterviews,thequestionsaskedofrespondentschange;hence,thetypeofinformationobtainedfromthosewhoareinterviewedatthebeginningmaybemarkedlydifferent from thatobtained from those interviewed towards theend.Also, this freedomcanintroduce investigatorbias into thestudy.Usingan interviewguideasameansofdatacollectionrequiresmuchmoreskillonthepartoftheresearcherthandoesusingastructuredinterview.
Constructingaresearchinstrumentinqualitativeresearch
Data inqualitative researcharenotcollected througha setofpredeterminedquestionsbutby raisingissuesarounddifferentareasofenquiry.Hence therearenopredeterminedresearch tools,assuch, inqualitativeresearch.However,manypeopledevelopalooselistofissuesthattheywanttodiscusswithrespondentsortohavereadyincasewhattheywanttodiscussdoesnotsurfaceduringthediscussions.Thislooselydevelopedlistofissuesiscalledaninterviewguide.Intheauthor’sopinion,particularlyforanewcomer,itisimportanttodevelopaninterviewguidetoensuredesiredcoverageoftheareasofenquiryandcomparabilityofinformationacrossrespondents.Notethatin-depthinterviewingisbothamethodofdatacollectionandastudydesigninqualitativeresearchandtheinterviewguideisaresearchtoolthatisusedtocollectdatainthisdesign.Recentlytheauthorconductedastudyusingin-depthinterviewingandfocusgroupmethodologiesto
construct a conceptual service deliverymodel for providing child protection services through familyconsultation, involvementandengagement.Theprojectwasdesigned todevelopamodel thatcanbeusedbythefieldworkerswhendealingwithafamilyonmattersrelatingtochildprotection.Theauthorconducted a number of in-depth interviews with some staff members working at different levels togatherideasoftheissuesthatserviceprovidersandmanagersthoughttobeimportant.Onthebasisoftheinformationobtainedfromthesein-depthinterviews,alistoflikelytopics/issueswasprepared.Thislist, theinterviewguide,becamethebasisofcollectingtherequiredinformationfromindividualsandfocus groups in order to construct the conceptual model. Though this list was developed the focusgroups were encouraged to raise any issue relating to the service delivery. The followingtopics/issues/questionsformedthecoreoftheinterviewguideforfocusgroups:
1. Whatdoyouunderstandbytheconceptoffamilyengagementandinvolvementwhendecidingaboutachild?
2. Whatshouldbetheextentandnatureoftheinvolvement?3. Howcanitbeachieved?4. Whatdoyouthinkaretheadvantagesofinvolvingfamiliesinthedecisionmaking?5. Whatinyouropinionareitsdisadvantages?6. Whatisyouropinionaboutthisconcept?7. Whatcanafieldworkerdotoinvolveafamily?8. Howcanthesuccessorfailureofthismodelbemeasured?9. Howwillthismodelaffectcurrentservicestochildren?
Note that these served as starting points for discussions. The groupmemberswere encouraged todiscusswhatevertheywantedtoinrelationtotheperceivedmodel.Allone-to-onein-depthinterviewsand focusgroupdiscussionswere recordedonaudiotapeandwereanalysed to identifymajor themesthatemergedfromthesediscussions.
Collectingdatausingsecondarysources
Sofarwehavediscussedtheprimarysourcesofdatacollectionwheretherequireddatawascollectedeitherbyyouorbysomeoneelseforthespecificpurposeyouhaveinmind.Thereareoccasionswhenyour data have already been collected by someone else and you need only to extract the requiredinformationforthepurposeofyourstudy.Both qualitative and quantitative research studies use secondary sources as a method of data
collection.Inqualitativeresearchyouusuallyextractdescriptive(historicalandcurrent)andnarrativeinformation and in quantitative research the information extracted is categorical or numerical. Thefollowingsectionprovidessomeofthemanysecondarysourcesgroupedintocategories:
Governmentorsemi-governmentpublications–Therearemanygovernmentandsemi-governmentorganisationsthatcollectdataonaregularbasisinavarietyofareasandpublishitforusebymembersofthepublicandinterestgroups.Somecommonexamplesarethecensus,vitalstatisticsregistration,labourforcesurveys,healthreports,economicforecastsanddemographicinformation.Earlierresearch–Forsometopics,anenormousnumberofresearchstudiesthathavealreadybeendonebyotherscanprovideyouwiththerequiredinformation.Personalrecords–Somepeoplewritehistoricalandpersonalrecords(e.g.diaries)thatmayprovidetheinformationyouneed.Massmedia–Reportspublishedinnewspapers,inmagazines,ontheInternet,andsoon,maybeanothergoodsourceofdata.
Problemswithusingdatafromsecondarysources
Whenusingdatafromsecondarysourcesyouneedtobecarefulastheremaybecertainproblemswiththeavailability,formatandqualityofdata.Theextentoftheseproblemsvariesfromsourcetosource.Whileusingsuchdatasomeissuesyoushouldkeepinmindare:
Validityandreliability–Thevalidityofinformationmayvarymarkedlyfromsourcetosource.Forexample,informationobtainedfromacensusislikelytobemorevalidandreliablethanthatobtainedfrommostpersonaldiaries.Personalbias–Theuseofinformationfrompersonaldiaries,newspapersandmagazinesmayhavetheproblemofpersonalbiasasthesewritersarelikelytoexhibitlessrigorousnessandobjectivitythanonewouldexpectinresearchreports.Availabilityofdata–Itiscommonforbeginningresearcherstoassumethattherequireddatawillbeavailable,butyoucannotandshouldnotmakethisassumption.Therefore,itisimportantto
makesurethattherequireddataisavailablebeforeyouproceedfurtherwithyourstudy.Format–Beforedecidingtousedatafromsecondarysourcesitisequallyimportanttoascertainthatthedataisavailableintherequiredformat.Forexample,youmightneedtoanalyseageinthecategories23–33,34–48,andsoon,but,inyoursource,agemaybecategorisedas21–24,25–29,andsoon.
SummaryIn thischapteryouhave learntabout thevariousmethodsofdatacollection. Informationcollectedaboutasituation,phenomenon,issueorgroupofpeoplecancomefromeitherprimarysourcesorsecondarysources.Primarysourcesarethosewhereyouorsomeoneelsecollectsinformationfromrespondentsforthespecificpurposeforwhicha
studyisundertaken.Theseincludeinterviewing,observationandtheuseofquestionnaires.Allothersources,wheretheinformationrequiredisalreadyavailable,suchasgovernmentpublications,reportsandpreviousresearch,arecalledsecondarysources.There is a considerable overlap in the methods of data collection between quantitative and qualitative research studies. The
differenceliesinthewaytheinformationisgenerated,recordedandanalysed.Inquantitativeresearchtheinformation,inmostcases,isgeneratedthroughasetofpredeterminedquestionsandeithertheresponsesarerecordedincategoricalformatorthecategoriesaredevelopedoutoftheresponses.Theinformationobtainedthengoesthroughdataprocessingandissubjectedtoanumberofstatisticalprocedures.Inqualitativeresearchtherequiredinformationisgeneratedthroughaseriesofquestionswhicharenotpredeterminedand pre-worded. In addition, the recording of information is in descriptive format and the dominantmode of analysis is contentanalysistoidentifythemainthemes.Structuredinterviews,useofquestionnairesandstructuredobservationsarethemostcommonmethodsofdatacollection inquantitativeresearch,whereas inqualitativeresearchunstructured interviews(oralhistories, in-depthinterviewsandnarratives)andparticipantobservationarethemainmethodsofdatacollectionfromprimarysources.Thechoiceofaparticularmethodofcollectingdatadependsuponthepurposeofcollectinginformation,thetypeofinformation
beingcollected,theresourcesavailabletoyou,yourskillsintheuseofaparticularmethodofdatacollectionandthesocioeconomic–demographiccharacteristicsofyourstudypopulation.Eachmethodhasitsownadvantagesanddisadvantagesandeachisappropriatefor certain situations. The choice of a particular method for collecting data is important in itself for ensuring the quality of theinformationbutnomethodofdatacollectionwillguarantee100percentaccurate information.Thequalityofyour information isdependent upon several methodological, situational and respondent-related factors and your ability as a researcher lies in eithercontrollingorminimisingtheeffectofthesefactorsintheprocessofdatacollection.Theuseofopen-endedandclosedquestionsisappropriatefordifferentsituations.Bothofthemhavestrengthsandweaknessesand
youshouldbeawareofthesesothatyoucanusethemappropriately.Theconstructionofaresearchinstrumentisthemostimportantaspectofanyresearchendeavourasitdeterminesthenatureand
qualityoftheinformation.Thisistheinputofyourstudyandtheoutput,therelevanceandaccuracyofyourconclusions,isentirelydependentupon it.A research instrument inquantitative researchmustbedeveloped in lightof theobjectivesofyour study.Themethod suggested in this chapter ensures that questions in an instrument have a direct link to your objectives. The wording ofquestionscanposeseveralproblemsandyoushouldkeeptheminmindwhileformulatingyourquestions.In qualitative research you do not develop a research instrument as such but it is advisable that you develop a conceptual
frameworkofthelikelyareasyouplantocover,providingsufficientallowancefornewonestoemergewhencollectingdatafromyourrespondents.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Identifytwoorthreeexamplesfromyourownacademicfieldwhereitmaybebettertouseaquestionnaireratherthaninterviewing,andviceversa.Identifythreesituationswhereitwouldbebettertouseopen-endedquestionsandthreewhereclosedquestionsmightbemoreuseful.Thereisaconsiderableoverlapinthemethodsofdatacollectionbetweenquantitativeandqualitativeresearch.Inspiteofthattheyaredifferent.Makealistofafewofthefactorsthatdifferentiatethem.
CHAPTER10CollectingDataUsingAttitudinalScales
Inthischapteryouwilllearnabout:
WhatattitudinalscalesareandhowtousethemThefunctionsofattitudinalscalesinquantitativeresearchDifficultiesindevelopinganattitudinalscaleandhowtoovercomethemDifferenttypesofattitudinalscalesandwhentousethemTherelationshipbetweenattitudinalandmeasurementscalesMethodsforexploringattitudesinqualitativeresearch
Keywords:attitudinalscales,attitudinalscore,attitudinalvalue,attitudinalweight,cumulativescale,equal-appearingscale,Guttmanscale,intervalscale,Likertscale,negative statements, neutral items,non-discriminate items,numerical scale, ordinalscale,positivestatements,ratioscale,summatedratingscale,Thurstonescale.
Measurementofattitudesinquantitativeandqualitativeresearch
There are a number of differences in the way attitudes aremeasured in quantitative and qualitativeresearch.Inquantitativeresearchyouareabletoexplore,measure,determinetheintensityandcombineattitudestodifferentaspectsofanissuetoarriveatoneindicatorthatisreflectiveoftheoverallattitude.Inqualitativeresearch,youcanonlyexplorethespreadofattitudesandestablishthetypesofattitudesprevalent.Inquantitativeresearchyoucanascertainthetypesofattitudespeoplehaveinacommunity,howmanypeoplehaveaparticularattitudeandwhat the intensity isof thoseattitudes.Anumberoftechniques have been developed tomeasure attitudes and their intensity in quantitative research, butsuchtechniquesarelackinginqualitativeresearch.Thisismainlybecauseinqualitativeresearchyoudo notmake an attempt tomeasure or quantify. The concept of attitudinal scales, therefore, is onlyprevalentinquantitativeresearch.
Attitudinalscalesinquantitativeresearch
Inquantitativeresearchtherearethreescaleswhichhavebeendevelopedto‘measure’attitudes.Eachofthesescalesisbasedupondifferentassumptionsandfollowsdifferentproceduresintheirconstruction.As a beginner in research methods it is important for you to understand these procedures and theassumptionsbehindthemsothatyoucanmakeappropriateandaccurateinterpretationofthefindings.Asyouwillsee,itisnotveryeasytoconstructanattitudinalscale.Outofthethreescales,theLikertscaleistheeasiesttoconstructandthereforeisusedfarmore.
Functionsofattitudinalscales
Ifyouwanttofindouttheattitudeofrespondentstowardsanissue,youcanaskeitheraclosedoranopen-endedquestion.For example, let us say that youwant to ascertain the attitudeof students in aclasstowardstheirlecturerandthatyouhaveaskedthemtorespondtothefollowingquestion:‘Whatisyour attitude towards your lecturer?’ If your question is open ended, it invites each respondent todescribe the attitude that s/heholds towards the lecturer. If youhave framed a closedquestion,withcategoriessuchas‘extremelypositive’,‘positive’,‘uncertain’,‘negative’and‘extremelynegative’,thisguidestherespondents toselectacategorythatbestdescribestheirattitude.This typeofquestioning,whether frameddescriptivelyor in a categorical form, elicits anoverall attitude towards the lecturer.Whileascertainingtheoverallattitudemaybesufficientinsomesituations,inmanyothers,wherethepurposeofattitudinalquestioningistodevelopstrategiesforimprovingaserviceorintervention,ortoformulatepolicy,elicitingattitudesonvariousaspectsoftheissueunderstudyisrequired.Butasyouknow,every issue, including thatof theattitudeofstudents towards their lecturers,has
many aspects. For example, the attitude of themembers of a community towards the provision of aparticularservicecomprises theirattitude towards theneed for theservice, itsmannerofdelivery, itslocation,thephysicalfacilitiesprovidedtousers,thebehaviourofthestaff,thecompetenceofthestaff,the effectiveness and efficiency of the service, and so on. Similarly, other examples – such as theattitudeofemployeestowardsthemanagementoftheirorganisation,theattitudeofemployeestowardsoccupationalredeploymentandredundancy,theattitudeofnursestowardsdeathanddying,theattitudeofconsumerstowardsaparticularproduct,theattitudeofstudentstowardsalecturer,ortheattitudeofstafftowardsthestrategicplanfortheirorganisation–canbebrokendowninthesamemanner.Respondentsusuallyhavedifferentattitudestowardsdifferentaspects.Onlywhenyouascertainthe
attitudeofrespondentstoanissuebyformulatingaquestionforeachaspect,usingeitheropen-endedorclosedquestions,doyoufindouttheirattitudetowardseachaspect.Themainlimitationofthismethodis that it is difficult to draw any conclusion about the overall attitude of a respondent from theresponses. Take the earlier example, where you want to find out the attitude of students towards alecturer.Therearedifferentaspectsof teaching: thecontentsof lectures; theorganisationofmaterial;the lecturer’s ability to communicatematerial; the presentation and style; knowledge of the subject;responsiveness;punctuality;andsoon.Studentsmayrate the lecturerdifferentlyondifferentaspects.Thatis,thelecturermightbeconsideredextremelycompetentandknowledgeableinhis/hersubjectbutmay not be considered a good communicator by amajority of students. Further, studentsmay differmarkedly in theiropinionregardinganyoneaspectofa lecturer’s teaching.Somemightconsider thelecturertobeagoodcommunicatorandothersmightnot.Themainproblemis:howdowefindoutthe‘overall’attitudeofthestudentstowardsthelecturer?Inotherwords,howdowecombinetheresponsestodifferentaspectsofanyissuetocomeupwithoneindicatorthatisreflectiveofanoverallattitude?Attitudinalscalesplayanimportantroleinovercomingthisproblem.Attitudinal scalesmeasure the intensity of respondents’ attitudes towards the various aspects of a
situation or issue and provide techniques to combine the attitudes towards different aspects into one
overallindicator.Thisreducestheriskofanexpressionofopinionbyrespondentsbeinginfluencedbytheiropinionononlyoneortwoaspectsofthatsituationorissue.
Difficultiesindevelopinganattitudinalscale
Indevelopinganattitudinalscaletherearethreeproblems:
1. Whichaspectsofasituationorissueshouldbeincludedwhenseekingtomeasureanattitude?Forinstance,intheexamplecitedabove,whataspectsofteachingshouldbeincludedinascaletofindouttheattitudeofstudentstowardstheirlecturer?
2. Whatprocedureshouldbeadoptedforcombiningthedifferentaspectstoobtainanoverallpicture?3. Howcanoneensurethatascalereallyismeasuringwhatitissupposedtomeasure?
The first problem is extremely important as it largely determines the third problem: the extent towhichthestatementsondifferentaspectsarereflectiveofthemainissuelargelydeterminesthevalidityof thescale.Youcansolve the thirdproblembyensuringthatyourstatementsonthevariousaspectshavealogicallinkwiththemainissueunderstudy–thegreaterthelink, thehigherthevalidity.Thedifferent types of attitudinal scale (Likert, Thurstone andGuttman) provide an answer to the secondproblem.Theyguideyouastotheprocedureforcombiningtheattitudestowardsvariousaspectsofanissue, though thedegreeofdifficulty in following theprocedure for these scalesvaries fromscale toscale.
Typesofattitudinalscale
Therearethreemajortypesofattitudinalscale:
1. thesummatedratingscale,alsoknownastheLikertscale;2. theequal-appearingintervalscaleordifferentialscale,alsoknownastheThurstonescale;3. thecumulativescale,alsoknownastheGuttmanscale.
ThesummatedratingorLikertscale
Thesummatedratingscale,morecommonlyknownastheLikertscale,isbasedupontheassumptionthateachstatement/itemonthescalehasequalattitudinalvalue,‘importance’or‘weight’intermsofreflectinganattitudetowardstheissueinquestion.Thisassumptionisalsothemainlimitationofthisscale as statements on a scale seldom have equal attitudinal value. For instance, in the examples inFigures10.1 and10.2, ‘knowledge of subject’ is not as important in terms of the degree towhich itreflectstheattitudeofthestudentstowardsthelectureras‘haspublishedagreatdeal’or‘somestudentslike,somedonot’,but,ontheLikertscale,eachistreatedashavingthesame‘weight’.Astudentmaynotbothermuchaboutwhetheralecturerhaspublishedagreatdeal,butmaybemoreconcernedabout‘knowledgeofthesubject’,‘communicateswell’and‘knowshowtoteach’.
FIGURE10.1AnexampleofacategoricalscaleItisimportanttorememberthattheLikertscaledoesnotmeasureattitudeperse.Itdoeshelptoplace
differentrespondentsinrelationtoeachotherintermsoftheintensityoftheirattitudetowardsanissue:itshowsthestrengthofonerespondent’sviewinrelationtothatofanotherandnottheabsoluteattitude.
FIGURE10.2Anexampleofaseven-pointnumericalscale
FIGURE10.3Anexampleofascalewithstatementsreflectingvaryingdegreesofanattitude
ConsiderationsinconstructingaLikertscale
In developing a Likert scale, there are a number of things to consider. Firstly, decide whether theattitude tobemeasured is tobeclassified intoone-, two-or three-directionalcategories (i.e.whetheryouwanttodeterminepositive,negativeandneutralpositionsinthestudypopulation)withrespecttotheir attitude towards the issue under study.Next, considerwhether youwant to use categories or anumerical scale.This shoulddependuponwhether you think that your studypopulation can expressitself better on a numerical scale or in categories. The decision about the number of points or thenumberofcategoriesonacategoricalscaledependsuponhowfinelyyouwanttomeasuretheintensity
oftheattitudeinquestionandonthecapacityofthepopulationtomakefinedistinctions.Figure10.1shows a five-point categorical scale that is threedirectional andFigure10.2 illustrates a seven-pointnumericalscale that isonedirectional.Sometimesyoucanalsodevelopstatementsreflectingopinionabout an issue in varying degrees (Figure 10.3). In this instance a respondent is asked to select thestatementwhichbestdescribestheopinion.
FIGURE10.4TheprocedureforconstructingaLikertscale
TheprocedureforconstructingaLikertscale
Figure10.4showstheprocedureusedinconstructingaLikertscale.
Calculatingattitudinalscores
Supposeyouhavedevelopedaquestionnaire/interviewscheduletomeasuretheattitudesofaclassofstudentstowardstheirlecturerusingascalewithfivecategories.InFigure10.5,statement1isapositivestatement;hence,ifarespondentticks‘stronglyagree’,s/he
isassumed tohaveamorepositiveattitudeon this item thanapersonwho ticks ‘agree’.Thepersonwhoticks‘agree’hasamorepositiveattitudethanapersonwhoticks‘uncertain’,andsoon.Therefore,apersonwhoticks‘stronglyagree’hasthemostpositiveattitudecomparedwithalloftheotherswithdifferent responses. Hence, the person is given the highest score, 5, as there are only five responsecategories.Iftherewerefourcategoriesyoucouldassignascoreof4.Asamatteroffact,anyscorecanbeassignedaslongastheintensityoftheresponsepatternisreflectedinthescoreandthehighestscoreisassignedtotheresponsewiththehighestintensity.
FIGURE10.5Scoringpositiveandnegativestatements
FIGURE10.6CalculatinganattitudinalscoreStatement2isanegativestatement.Inthiscaseapersonwhoticks‘stronglydisagree’hasthemost
positiveattitudeonthisitem;hence,thehighestscoreisassigned,5.Ontheotherhand,arespondentwhoticks‘stronglyagree’hastheleastpositiveattitudeontheitemandthereforeisassignedthelowestscore,1.Thesamescoringsystemisfollowedfortheotherstatements.Notestatement9.Therewillalwaysbesomepeoplewholikealecturerandsomewhodonot;hence,
thistypeofstatementisneutral.Thereisnopointinincludingsuchitemsinthescalebut,here,forthepurposeofthisexample,wehave.To illustrate how to calculate an individual’s attitudinal score, let us take the example of two
respondentswhohavetickedthedifferentstatementsmarkedinourexampleby#and@(seeFigure10.6).Letusworkouttheirattitudinalscore:
The analysis shows that, overall, respondent@has a ‘more’ positive attitude towards the lecturerthanrespondent#.Youcannotsaythattheattitudeofrespondent@istwice(42/20=2.10)aspositiveas that of respondent #. The attitudinal score only places respondents in a position relative to oneanother.Remember that theLikert scaledoesnotmeasure theattitudeperse,buthelpsyou to rateagroupofindividualsindescendingorascendingorderwithrespecttotheirattitudestowardstheissuesinquestion.
Theequal-appearingintervalorThurstonescale
Unlike the Likert scale, the Thurstone scale calculates a ‘weight’ or ‘attitudinal value’ for eachstatement.Theweight(equivalenttothemedianvalue)foreachstatementiscalculatedonthebasisofratingassignedbyagroupofjudges.Eachstatementwithwhichrespondentsexpressagreement(ortowhichtheyrespondintheaffirmative)isgivenanattitudinalscoreequivalenttothe‘attitudinalvalue’ofthestatement.TheprocedureforconstructingtheThurstonescaleisasgiveninFigure10.7.
FIGURE10.7TheprocedureforconstructingtheThurstonescaleThemainadvantageofthisscaleisthat,astheimportanceofeachstatementisdeterminedbyjudges,
itreflectstheabsoluteratherthanrelativeattitudesofrespondents.Thescaleisthusabletoindicatetheintensityofpeople’sattitudesandanychange in this intensity should the studybe replicated.On theotherhand,thescaleisdifficulttoconstruct,andamajorcriticismisthatjudgesandrespondentsmayassess the importance of a particular statement differently and, therefore, the respondents’ attitudesmightnotbereflected.
ThecumulativeorGuttmanscale
TheGuttmanscale is oneof themostdifficult scales to construct and therefore is rarelyused.Thisscaledoesnothavemuchrelevanceforbeginnersinresearchandsoisnotdiscussedinthisbook.
Attitudinalscalesandmeasurementscales
Differentattitudinalscalesusedifferentmeasurementscales.Itisimportanttoknowwhichattitudinalscale belongs towhichmeasurement scale as thiswill help you in the interpretation of respondents’scores.Table10.1showsattitudinalscalesinrelationtomeasurementscales.
TABLE10.1Therelationshipbetweenattitudinalandmeasurementscales
Attitudinalscales Measurementscales
Likertscale OrdinalscaleThurstonescale IntervalscaleGuttmanscale Ratioscale
Attitudesandqualitativeresearch
Asmentionedatthebeginningofthischapter,inqualitativeresearchyoucanonlyexplorethespreadofthe attitudes. Whatever methods of data collection you use – in-depth interviewing, focus group,
observation–youcanexplorethediversityintheattitudesbutcannotfindotheraspectslike:howmanypeoplehaveaparticularattitude,theintensityofaparticularattitude,oroverallwhattheattitudeofapersonis.Qualitativemethodsarethereforebestsuitedtoexplorethediversityinattitudes.
SummaryOneof thesignificantdifferencesbetweenquantitativeandqualitativeresearch is in theavailabilityofmethodsandprocedures tomeasure attitudes. In quantitative research there are a number of methods that can be used to measure attitudes but qualitativeresearch lacks methodology in this aspect primarily because its aim is to explain rather than to measure and quantify. Throughqualitative researchmethodology you can find the diversity or spread of attitudes towards an issue but not their intensity and acombinedoverallindicator.Attitudinal scales are used in quantitative research tomeasure attitudes towards an issue. Their strength lies in their ability to
combineattitudestowardsdifferentaspectsofanissueandtoprovideanindicatorthatisreflectiveofanoverallattitude.However,thereareproblemsindevelopinganattitudinalscale.Youmustdecidewhichaspectsshouldbeincludedwhenmeasuringattitudestowardsan issue,howtheresponsesgivenbya respondentshouldbecombined toascertain theoverallattitude,andhowyoucanensurethatthescaledevelopedreallymeasuresattitudetowardstheissueinquestion.Therearethreetypesofscalethatmeasureattitude:theLikert,ThurstoneandGuttmanscales.TheLikertscaleismostcommonly
usedbecauseitiseasytoconstruct.Themainassumptionofthescaleisthateachstatementis‘equallyimportant’.The‘importance’ofeachitemfortheThurstonescaleisdeterminedbyapanelofjudges.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.IdentifyexamplesofhowtheLikertandThurstonescalescanbeappliedtoresearchinyourownacademicfield.Considerhowyouwouldgoaboutdevelopingafive-pointLikertscaletomeasuretheself-esteemofagroupofuniversitystudents,andthedifficultiesyoumightfaceintryingtodoso.
CHAPTER11EstablishingtheValidityandReliabilityofaResearchInstrument
Inthischapteryouwilllearnabout:
TheconceptofvalidityDifferenttypesofvalidityinquantitativeresearchTheconceptofreliabilityFactorsaffectingthereliabilityofaresearchinstrumentMethodsofdeterminingthereliabilityofaninstrumentinquantitativeresearchValidityandreliabilityinqualitativeresearch
Keywords: concurrent validity, confirmability, construct validity, content validity,credibility, dependability, external consistency, face validity, internal consistency,reliability,transferability,validity.
Intheprevioustwochapterswediscussedvariousmethodsofdatacollectioninbothquantitativeandqualitative research. The questions asked of your respondents are the basis of your findings andconclusions.Thesequestionsconstitutethe‘input’foryourconclusions(the‘output’).Thisinputpassesthrougha seriesof steps– the selectionof a sample, thecollectionof information, theprocessingofdata,theapplicationofstatisticalproceduresandthewritingofareport–andthemannerinwhichallofthesearedonecanaffecttheaccuracyandqualityofyourconclusions.Hence,itisimportantforyoutoattempt to establish the quality of your results. As a researcher you can also be asked by others toestablishtheappropriateness,qualityandaccuracyoftheproceduresyouadoptedforfindinganswerstoyourresearchquestions.Broadly,thisconceptofappropriatenessandaccuracyasappliedtoaresearchprocessiscalledvalidity.Asinaccuraciescanbeintroducedintoastudyatanystage, theconceptofvaliditycanbeappliedtotheresearchprocessasawholeortoanyofitssteps:studydesign,samplingstrategy, conclusions drawn, the statistical procedures applied or the measurement procedures used.Broadly,therearetwoperspectivesonvalidity:
1. Istheresearchinvestigationprovidinganswerstotheresearchquestionsforwhichitwasundertaken?
2. Ifso,isitprovidingtheseanswersusingappropriatemethodsandprocedures?
Inthischapterwewilldiscusstheconceptofvalidityasappliedtomeasurementproceduresortheresearchtoolsusedtocollecttherequiredinformationfromyourrespondents.There are prominent differences between quantitative and qualitative research in relation to the
conceptsofvalidityand reliability.Becauseof thedefinedandestablishedstructuresandmethodsofdata collection in quantitative research, the concepts of validity and reliability and the methods todeterminethemarewelldeveloped.However,thesameisnotthecaseinqualitativeresearchwhereitwouldbeappropriatetosaythattheseconceptscannotberigorouslyappliedinthesamewayastheyareinquantitativeresearchbecauseoftheflexibility,freedomandspontaneitygiventoaresearcherinthemethodsandproceduresofdatacollection. Itbecomesdifficult toestablishstandardisation in themethod(s) of data collection in qualitative research and, hence, their validity and reliability. Despitethesedifficultiestherearesomemethodswhichhavebeenproposedtoestablishvalidityandreliabilityinqualitativeresearchwhicharedetailedinthischapter.
Theconceptofvalidity
Toexamine theconceptofvalidity, letus takeaverysimpleexample.Supposeyouhavedesignedastudy to ascertain the health needs of a community. In doing so, you have developed an interviewschedule.Furthersupposethatmostofthequestionsintheinterviewschedulerelatetotheattitudeofthestudypopulationtowardsthehealthservicesbeingprovidedtothem.Notethatyouraimwastofindoutabouthealthneedsbuttheinterviewscheduleisfindingoutwhatattitudesrespondentshavetothehealthservices;thus,theinstrumentisnotmeasuringwhatitwasdesignedtomeasure.Theauthorhascomeacrossmanysimilarexamplesamongstudentsandlessskilledresearchers.In terms ofmeasurement procedures, therefore, validity is the ability of an instrument tomeasure
whatitisdesignedtomeasure:‘Validityisdefinedasthedegreetowhichtheresearcherhasmeasuredwhathehassetouttomeasure’(Smith1991:106).AccordingtoKerlinger,‘Thecommonestdefinitionofvalidityisepitomisedbythequestion:Arewemeasuringwhatwethinkwearemeasuring?’(1973:457).Babbiewrites,‘validityreferstotheextenttowhichanempiricalmeasureadequatelyreflectstherealmeaningoftheconceptunderconsideration’(1989:133).Thesedefinitionsraisetwokeyquestions:
Whodecideswhetheraninstrumentismeasuringwhatitissupposedtomeasure?Howcanitbeestablishedthataninstrumentismeasuringwhatitissupposedtomeasure?
Obviouslytheanswertothefirstquestionisthepersonwhodesignedthestudy,thereadershipofthereportandexpertsinthefield.Thesecondquestionisextremelyimportant.Onwhatbasisdoyou(asaresearcher), a reader as a consumer or an expert make this judgement? In the social sciences thereappeartobetwoapproachestoestablishingthevalidityofaresearchinstrument.Theseapproachesarebaseduponeitherlogicthatunderpinstheconstructionoftheresearchtoolorstatisticalevidencethatisgatheredusing informationgenerated through theuseof the instrument.Establishingvalidity throughlogic implies justification of each question in relation to the objectives of the study, whereas thestatistical procedures provide hard evidence by way of calculating the coefficient of correlationsbetweenthequestionsandtheoutcomevariables.Establishingalogicallinkbetweenthequestionsandtheobjectivesisbothsimpleanddifficult.Itis
simple in the sense that youmay find it easy to see a link for yourself, and difficult because yourjustificationmaylackthebackingofexpertsandthestatisticalevidencetoconvinceothers.Establishingalogicallinkbetweenquestionsandobjectivesiseasierwhenthequestionsrelatetotangiblematters.
For example, if you want to find out about age, income, height or weight, it is relatively easy toestablishthevalidityofthequestions,buttoestablishwhetherasetofquestionsismeasuring,say,theeffectiveness of a programme, the attitudes of a group of people towards an issue, or the extent ofsatisfaction of a group of consumerswith the service provided by an organisation ismore difficult.Whenalesstangibleconceptisinvolved,suchaseffectiveness,attitudeorsatisfaction,youneedtoaskseveralquestionsinordertocoverdifferentaspectsoftheconceptanddemonstratethatthequestionsasked are actuallymeasuring it. Validity in such situations becomesmore difficult to establish, andespecially in qualitative researchwhere you aremostly exploring feelings, experiences, perceptions,motivationsorstories.Itisimportanttorememberthattheconceptofvalidityispertinentonlytoaparticularinstrumentand
itisanidealstatethatyouasaresearcheraimtoachieve.
Typesofvalidityinquantitativeresearch
Therearethreetypesofvalidityinquantitativeresearch:
1. faceandcontentvalidity;2. concurrentandpredictivevalidity;3. constructvalidity.
Faceandcontentvalidity
The judgement that an instrument is measuring what it is supposed to is primarily based upon thelogicallinkbetweenthequestionsandtheobjectivesofthestudy.Hence,oneofthemainadvantagesofthis typeofvalidity is that it iseasy toapply.Eachquestionor itemon theresearch instrumentmusthave a logical linkwith an objective. Establishment of this link is called facevalidity. It is equallyimportant that the items and questions cover the full range of the issue or attitude beingmeasured.Assessment of the items of an instrument in this respect is called content validity. In addition, thecoverage of the issue or attitude should be balanced; that is, each aspect should have similar andadequate representation in the questions or items.Content validity is also judgedon the basis of theextenttowhichstatementsorquestionsrepresenttheissuetheyaresupposedtomeasure,asjudgedbyyou as a researcher, your readership and experts in the field. Although it is easy to present logicalargumentstoestablishvalidity,therearecertainproblems:
Thejudgementisbaseduponsubjectivelogic;hence,nodefiniteconclusionscanbedrawn.Differentpeoplemayhavedifferentopinionsaboutthefaceandcontentvalidityofaninstrument.Theextenttowhichquestionsreflecttheobjectivesofastudymaydiffer.Iftheresearchersubstitutesonequestionforanother,themagnitudeofthelinkmaybealtered.Hence,thevalidityoritsextentmayvarywiththequestionsselectedforaninstrument.
Concurrentandpredictivevalidity
‘In situations where a scale is developed as an indicator of some observable criterion, the scale’svalidity can be investigated by seeing how good an indicator it is’ (Moser & Kalton 1989: 356).
Suppose you develop an instrument to determine the suitability of applicants for a profession. Theinstrument’svaliditymightbedeterminedbycomparingitwithanotherassessment,forexamplebyapsychologist,orwitha futureobservationofhowwell theseapplicantshavedone in the job. Ifbothassessmentsaresimilar,theinstrumentusedtomaketheassessmentatthetimeofselectionisassumedtohavehighervalidity.Thesetypesofcomparisonsestablishtwotypesofvalidity:predictivevalidityandconcurrentvalidity.Predictivevalidityisjudgedbythedegreetowhichaninstrumentcanforecastan outcome. Concurrent validity is judged by how well an instrument compares with a secondassessment concurrently done: ‘It is usually possible to express predictive validity in terms of thecorrelation coefficient between the predicted status and the criterion. Such a coefficient is called avaliditycoefficient’(Burns1997:220).
Constructvalidity
Constructvalidityisamoresophisticatedtechniqueforestablishingthevalidityofaninstrument.Itisbaseduponstatisticalprocedures.Itisdeterminedbyascertainingthecontributionofeachconstructtothetotalvarianceobservedinaphenomenon.Supposeyouare interested incarryingoutastudy tofind thedegreeof jobsatisfactionamongthe
employeesofanorganisation.Youconsiderstatus,thenatureofthejobandremunerationasthethreemostimportantfactorsindicativeofjobsatisfaction,andconstructquestionstoascertainthedegreetowhichpeopleconsidereachfactorimportantforjobsatisfaction.Afterthepre-testordataanalysisyouusestatisticalprocedurestoestablishthecontributionofeachconstruct(status,thenatureofthejobandremuneration) to the total variance (job satisfaction). The contribution of these factors to the totalvarianceisanindicationofthedegreeofvalidityoftheinstrument.Thegreaterthevarianceattributabletotheconstructs,thehigherthevalidityoftheinstrument.One of themain disadvantages of construct validity is that you need to know about the required
statisticalprocedures.
Theconceptofreliability
Weusetheword‘reliable’veryofteninourlives.Whenwesaythatapersonisreliable,whatdowemean?Weinferthats/heisdependable,consistent,predictable,stableandhonest.Theconceptofreliabilityinrelationtoaresearchinstrumenthasasimilarmeaning:ifaresearchtool
isconsistentandstable,hencepredictableandaccurate,itissaidtobereliable.Thegreaterthedegreeof consistency and stability in an instrument, the greater its reliability. Therefore, ‘a scale or test isreliabletotheextentthatrepeatmeasurementsmadebyitunderconstantconditionswillgivethesameresult’(Moser&Kalton1989:353).Theconceptofreliabilitycanbelookedatfromtwosides:
1. Howreliableisaninstrument?2. Howunreliableisit?
Thefirstquestionfocusesontheabilityofaninstrumenttoproduceconsistentmeasurements.Whenyoucollectthesamesetofinformationmorethanonceusingthesameinstrumentandgetthesameorsimilar results under the same or similar conditions, an instrument is considered to be reliable. Thesecondquestionfocusesonthedegreeofinconsistencyinthemeasurementsmadebyaninstrument–
thatis,theextentofdifferenceinthemeasurementswhenyoucollectthesamesetofinformationmorethan once, using the same instrument under the same or similar conditions. Hence, the degree ofinconsistencyinthedifferentmeasurementsisanindicationoftheextentofitsinaccuracy.This‘error’is a reflection of an instrument’s unreliability. Therefore, reliability is the degree of accuracy orprecision in themeasurementsmade by a research instrument.The lower the degree of ‘error’ in aninstrument,thehigherthereliability.Letustakeanexample.Supposeyoudevelopaquestionnairetoascertaintheprevalenceofdomestic
violenceinacommunity.Youadministerthisquestionnaireandfindthatdomesticviolenceisprevalentin,say,5percentofhouseholds.Ifyoufollowthiswithanothersurveyusingthesamequestionnaireonthesamepopulationunderthesameconditions,anddiscoverthattheprevalenceofdomesticviolenceis, say, 15 per cent, the questionnaire has not given a comparable result, which may mean it isunreliable. The less the difference between the two sets of results, the higher the reliability of theinstrument.
Factorsaffectingthereliabilityofaresearchinstrument
Inthesocialsciencesit isimpossibletohavearesearchtoolwhichis100percentaccurate,notonlybecause a research instrument cannot be so, but also because it is impossible to control the factorsaffectingreliability.Someofthesefactorsare:
Thewordingofquestions–Aslightambiguityinthewordingofquestionsorstatementscanaffectthereliabilityofaresearchinstrumentasrespondentsmayinterpretthequestionsdifferentlyatdifferenttimes,resultingindifferentresponses.Thephysicalsetting–Inthecaseofaninstrumentbeingusedinaninterview,anychangeinthephysicalsettingatthetimeoftherepeatinterviewmayaffecttheresponsesgivenbyarespondent,whichmayaffectreliability.Therespondent’smood–Achangeinarespondent’smoodwhenrespondingtoquestionsorwritinganswersinaquestionnairecanchangeandmayaffectthereliabilityofthatinstrument.Theinterviewer’smood–Asthemoodofarespondentcouldchangefromoneinterviewtoanothersocouldthemood,motivationandinteractionoftheinterviewer,whichcouldaffecttheresponsesgivenbyrespondentstherebyaffectingthereliabilityoftheresearchinstrument.Thenatureofinteraction–Inaninterviewsituation,theinteractionbetweentheinterviewerandtheintervieweecanaffectresponsessignificantly.Duringtherepeatinterviewtheresponsesgivenmaybedifferentduetoachangeininteraction,whichcouldaffectreliability.Theregressioneffectofaninstrument–Whenaresearchinstrumentisusedtomeasureattitudestowardsanissue,somerespondents,afterhavingexpressedtheiropinion,mayfeelthattheyhavebeeneithertoonegativeortoopositivetowardstheissue.Thesecondtimetheymayexpresstheiropiniondifferently,therebyaffectingreliability.
Methodsofdeterminingthereliabilityofaninstrumentinquantitativeresearch
Thereareanumberofwaysofdeterminingthereliabilityofaninstrumentandthesecanbeclassifiedaseitherexternalorinternalconsistencyprocedures.
Externalconsistencyprocedures
External consistencyprocedurescompare findings from two independentprocessesofdatacollectionwitheachotherasameansofverifyingthereliabilityofthemeasure.Thetwomethodsofdoingthisareasfollows:
1. Test/retest–Thisisacommonlyusedmethodforestablishingthereliabilityofaresearchtool.Inthetest/retest(repeatabilitytest)aninstrumentisadministeredonce,andthenagain,underthesameorsimilarconditions.Theratiobetweenthetestandretestscores(oranyotherfinding,forexampletheprevalenceofdomesticviolence,adiseaseorincidenceofanillness)isanindicationofthereliabilityoftheinstrument–thegreaterthevalueoftheratio,thehigherthereliabilityoftheinstrument.Asanequation,
(testscore)/(retest)=1
or
(testscore)–(retest)=0
Aratioof1shows100percentreliability(nodifferencebetweentestandretest)andanydeviationfrom it indicates less reliability – the less the value of this ratio, the less the reliability of theinstrument. Expressed in another way, zero difference between the test and retest scores is anindication of 100 per cent reliability. The greater the difference between scores or findingsobtainedfromthetwotests,thegreatertheunreliabilityoftheinstrument.Themainadvantageofthetest/retestprocedureisthatitpermitstheinstrumenttobecompared
withitself,thusavoidingthesortofproblemsthatcouldarisewiththeuseofanotherinstrument.Themaindisadvantageof thismethod is that a respondentmay recall the responses that s/he
gave in the first round, which in turn may affect the reliability of the instrument. Where aninstrumentisreactiveinnature(whenaninstrumenteducatestherespondentwithrespecttowhatthe researcher is trying to find out) thismethodwill not provide an accurate assessment of itsreliability.Oneof thewaysofovercomingthisproblemis to increasethetimespanbetweenthetwo tests,but thismayaffect reliability forother reasons, suchas thematurationof respondentsandtheimpossibilityofachievingconditionssimilar to thoseunderwhichthequestionnairewasfirstadministered.
2. Parallelformsofthesametest–Inthisprocedureyouconstructtwoinstrumentsthatareintendedtomeasurethesamephenomenon.Thetwoinstrumentsarethenadministeredtotwosimilarpopulations.Theresultsobtainedfromonetestarecomparedwiththoseobtainedfromtheother.Iftheyaresimilar,itisassumedthattheinstrumentisreliable.Themainadvantageofthisprocedureisthatitdoesnotsufferfromtheproblemofrecallfoundinthetest/retestprocedure.Also,atimelapsebetweenthetwotestsisnotrequired.Adisadvantageisthatyouneedtoconstructtwoinstrumentsinsteadofone.Moreover,itisextremelydifficulttoconstructtwoinstrumentsthatarecomparableintheirmeasurementofaphenomenon.Itisequallydifficulttoachievecomparabilityinthetwopopulationgroupsandinthetwoconditionsunderwhichthetestsareadministered.
Internalconsistencyprocedures
The idea behind internal consistency procedures is that items or questions measuring the samephenomenon,iftheyarereliableindicators,shouldproducesimilarresultsirrespectiveoftheirnumberinaninstrument.Evenifyourandomlyselectafewitemsorquestionsoutofthetotalpooltotestthereliabilityofaninstrument,eachsegmentofquestionsthusconstructedshouldreflectreliabilitymoreorlesstothesameextent.Itisbaseduponthelogicthatifeachitemorquestionisanindicatorofsomeaspectofaphenomenon,eachsegmentconstructedwillstillreflectdifferentaspectsofthephenomenoneventhoughitisbaseduponfeweritems/questions.Hence,evenifwereducethenumberofitemsorquestions,aslongastheyreflectsomeaspectofaphenomenon,alessernumberofitemscanprovideanindication of the reliability of an instrument. The internal consistency procedure is based upon thislogic.The followingmethod is commonlyused formeasuring the reliability of an instrument in thisway:
Thesplit-halftechnique–Thistechniqueisdesignedtocorrelatehalfoftheitemswiththeotherhalfandisappropriateforinstrumentsthataredesignedtomeasureattitudestowardsanissueorphenomenon.Thequestionsorstatementsaredividedinhalfinsuchawaythatanytwoquestionsorstatementsintendedtomeasurethesameaspectfallintodifferenthalves.Thescoresobtainedbyadministeringthetwohalvesarecorrelated.Reliabilityiscalculatedbyusingtheproductmomentcorrelation(astatisticalprocedure)betweenscoresobtainedfromthetwohalves.Becausetheproductmomentcorrelationiscalculatedonthebasisofonlyhalftheinstrument,itneedstobecorrectedtoassessreliabilityforthewhole.Thisisknownasstepped-upreliability.Thestepped-upreliabilityforthewholeinstrumentiscalculatedbyaformulacalledtheSpearman–Brownformula(astatisticalprocedure).
Validityandreliabilityinqualitativeresearch
One of the areas of difference between quantitative and qualitative research is in the use of and theimportancegiventotheconceptsofvalidityandreliability.Thedebatecentresonwhetherornot,giventhe framework of qualitative research, these concepts can or even should be applied in qualitativeresearch.Asyouknow, validity in the broader sense refers to the ability of a research instrument todemonstrate that it is finding out what you designed it to and reliability refers to consistency in itsfindingswhenused repeatedly. Inqualitative research, as answers to researchquestions are exploredthrough multiple methods and procedures which are both flexible and evolving, to ensurestandardisationofresearchtoolsaswellastheprocessesbecomesdifficult.Asanewcomertoresearchyou may wonder how these concepts can be applied in qualitative research when it does not usestandardised and structured methods and procedures which are the bases of testing validity andreliability as defined in quantitative research. Youmay ask how you can ascertain the ability of aninstrumenttomeasurewhatitisexpectedtoandhowconsistentitiswhenthedatacollectionquestionsareneitherfixednorstructured.However, there are some attempts to define and establish validity and reliability in qualitative
research. In a chapter entitled ‘Competing paradigms in qualitative research’ (pp. 105–117) in theHandbook of Qualitative Research, edited by Denzin and Lincoln (1994), Guba and Lincoln havesuggestedaframeworkoffourcriteriaasapartoftheconstructivismparadigmparalleling‘validity’and‘reliability’ inquantitative research.According to them, thereare twosetsofcriteria ‘for judging thegoodnessorqualityofaninquiryinconstructivismparadigm’(1994:114).Theseare:‘trustworthiness’and‘authenticity’.AccordingtoGubaandLincoln,trustworthinessinaqualitativestudyisdetermined
by four indicators – credibility, transferability,dependability and confirmability – and it is thesefourindicatorsthatreflectvalidityandreliabilityinqualitativeresearch.‘Thetrustworthinesscriteriaofcredibility (paralleling internal validity), transferability (paralleling external validity), dependability(paralleling reliability), and confirmability (paralleling objectivity)’, according to Guba and Lincoln(1994:114)closelyrelatestotheconceptsofvalidityandreliability.TrochimandDonnelly(2007)compare thecriteriaproposedbyGubaandLincolnin thefollowing
tablewithvalidityandreliabilityasdefinedinquantitativeresearch:Traditionalcriteriaforjudgingquantitativeresearch AlternativecriteriaforjudgingqualitativeresearchInternalValidity CredibilityExternalValidity TransferabilityReliability DependabilityObjectivity Confirmability
(TrochimandDonnelly2007:149)
Credibility–AccordingtoTrochimandDonnelly(2007:149),‘credibilityinvolvesestablishingthattheresultsofqualitativeresearcharecredibleorbelievablefromtheperspectiveoftheparticipantintheresearch’.Asqualitativeresearchstudiesexploreperceptions,experiences,feelingsandbeliefsofthepeople,itisbelievedthattherespondentsarethebestjudgetodeterminewhetherornottheresearchfindingshavebeenabletoreflecttheiropinionsandfeelingsaccurately.Hence,credibility,whichissynonymoustovalidityinquantitativeresearch,isjudgedbytheextentofrespondentconcordancewherebyyoutakeyourfindingstothosewhoparticipatedinyourresearchforconfirmation,congruence,validationandapproval.Thehighertheoutcomeofthese,thehigherthevalidityofthestudy.Transferability–This‘referstothedegreetowhichtheresultsofqualitativeresearchcanbegeneralizedortransferredtoothercontextsorsettings’(2007:149).Thoughitisverydifficulttoestablishtransferabilityprimarilybecauseoftheapproachyouadoptinqualitativeresearch,tosomeextentthiscanbeachievedifyouextensivelyandthoroughlydescribetheprocessyouadoptedforotherstofollowandreplicate.Dependability–IntheframeworksuggestedbyGubaandLincolnthisisverysimilartotheconceptofreliabilityinquantitativeresearch:‘Itisconcernedwithwhetherwewouldobtainthesameresultsifwecouldobservethesamethingtwice’(TrochimandDonnelly2007:149).Again,asqualitativeresearchadvocatesflexibilityandfreedom,itmaybedifficulttoestablishunlessyoukeepanextensiveanddetailedrecordoftheprocessforotherstoreplicatetoascertainthelevelofdependability.Confirmability–This‘referstothedegreetowhichtheresultscouldbeconfirmedorcorroboratedbyothers’(2007:149).Confirmabilityisalsosimilartoreliabilityinquantitativeresearch.Itisonlypossibleifbothresearchersfollowtheprocessinanidenticalmannerfortheresultstobecompared.
To the author’smind, to some extent, it is possible to establish the ‘validity’ and ‘reliability’ of thefindingsinqualitativeresearchintheformofthemodelsuggestedbyGubaandLincoln,butitssuccessismostlydependentupontheidenticalreplicationoftheprocessandmethodsfordatacollectionwhichmaynotbeeasytoachieveinqualitativeresearch.
SummaryOneofthedifferencesinquantitativeandqualitativeresearchisintheuseofandimportanceattachedtotheconceptsofvalidityandreliability.Theseconcepts,theiruseandmethodsofdeterminationaremoreacceptedanddevelopedinquantitativethanqualitativeresearch.Theconceptofvalidityreferstoasituationwherethefindingsofyourstudyareinaccordancewithwhatyoudesignedittofindout.Thenotionof validity canbe applied to any aspect of the researchprocess.With respect tomeasurement procedures, itrelatestowhetheraresearchinstrumentismeasuringwhatitsetouttomeasure.Inquantitativeresearch,therearetwoapproachesusedtoestablishthevalidityofaninstrument:theestablishmentofalogicallinkbetweentheobjectivesofastudyandthequestionsusedinaninstrument,andtheuseofstatisticalanalysistodemonstratetheselinks.Therearethreetypesofvalidityinquantitativeresearch:faceandcontent,concurrentandpredictive,andconstructvalidity.However,theuseoftheconceptofvalidityinqualitativeresearchisdebatableandcontroversial.Inqualitativeresearch‘credibility’asdescribedbyGubaandLincolnseemstobetheonlyindicatorofinternalvalidityandisjudgedbythedegreeofrespondentconcordancewiththefindings.Themethodsusedtoestablish‘validity’aredifferentinquantitativeandqualitativeresearch.The reliability of an instrument refers to its ability to produce consistent measurements each time. When we administer an
instrument under the same or similar conditions to the same or similar population and obtain similar results, we say that theinstrumentis‘reliable’–themoresimilartheresults,thegreaterthereliability.Youcanlookatreliabilityfromtwosides:reliability(theextentofaccuracy)andunreliability(theextentofinaccuracy).Ambiguityinthewordingofquestions,achangeinthephysicalsettingfordatacollection,arespondent’smoodwhenprovidinginformation, thenatureof the interactionbetweeninterviewerandinterviewee,andtheregressiveeffectofaninstrumentarefactorsthatcanaffectthereliabilityofaresearchinstrument.Inqualitativeresearch‘reliability’ismeasuredthrough‘dependability’and‘confirmability’assuggestedbyGubaandLincoln.Thereareexternalandinternalconsistencyproceduresfordeterminingreliabilityinquantitativeresearch.Test/retestandparallel
formsofthesametestarethetwoproceduresthatdeterminetheexternalreliabilityofaresearchinstrument,whereasthesplit-halftechnique is classified under internal consistency procedures. There seem to be no set procedures for determining the variousindicatorsofvalidityandreliabilityinqualitativeresearch.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Explorehowtheconceptsofreliabilityandvalidityareapplicabletoresearchinyouracademicfieldorprofession.Considerwhatstrategiesorproceduresyoucouldputinplacetolimittheaffectonreliabilityofthefollowingfactors:
wordingofquestions;physicalsetting;respondent’smood;interviewer’smood;natureofinteraction;regressioneffectofaninstrument.
STEPIVSelectingaSample
Thisoperationalstepincludesonechapter:
Chapter12:Selectingasample
CHAPTER12SelectingaSample
Inthischapteryouwilllearnabout:
ThedifferencesbetweensamplinginqualitativeandquantitativeresearchDefinitionsofsamplingterminologyThetheoreticalbasisforsamplingFactorsaffectingtheinferencesdrawnfromasampleDifferenttypesofsamplingincluding:
Random/probabilitysamplingdesignsNon-random/non-probabilitysamplingdesignsThe‘mixed’samplingdesign
ThecalculationofsamplesizeTheconceptofsaturationpoint
Keywords: accidental sampling, cluster sampling, data saturation point,disproportionate sampling, equal and independent, estimate, information-rich,judgementalsampling,multi-stageclustersampling,non-randomsample,populationmean, population parameters, quota sampling, random numbers, random sample,sample statistics, sampling, sampling design, sampling element, sampling error,samplingframe,samplingpopulation,samplingunit,samplesize,samplingstrategy,saturationpoint,snowballsampling,studypopulation,stratifiedsampling,systematicsampling.
Thedifferencesbetweensamplinginquantitativeandqualitativeresearch
The selection of a sample in quantitative and qualitative research is guided by two opposingphilosophies.Inquantitativeresearchyouattempttoselectasampleinsuchawaythat it isunbiasedandrepresentsthepopulationfromwhereitisselected.Inqualitativeresearch,numberconsiderations
mayinfluencetheselectionofasamplesuchas:theeaseinaccessingthepotentialrespondents;yourjudgementthatthepersonhasextensiveknowledgeaboutanepisode,aneventorasituationofinteresttoyou;howtypicalthecaseisofacategoryofindividualsorsimplythatitistotallydifferentfromtheothers.Youmakeeveryeffort toselecteitheracasethat issimilartotherestofthegrouportheonewhichistotallydifferent.Suchconsiderationsarenotacceptableinquantitativeresearch.Thepurposeofsamplinginquantitativeresearchistodrawinferencesaboutthegroupfromwhich
you have selected the sample, whereas in qualitative research it is designed either to gain in-depthknowledgeaboutasituation/event/episodeortoknowasmuchaspossibleaboutdifferentaspectsofanindividualontheassumptionthattheindividualistypicalofthegroupandhencewillprovideinsightintothegroup.Similarly,thedeterminationofsamplesizeinquantitativeandqualitativeresearchisbaseduponthe
twodifferentphilosophies.Inquantitativeresearchyouareguidedbyapredeterminedsamplesizethatis based upon a number of other considerations in addition to the resources available. However, inqualitativeresearchyoudonothaveapredeterminedsamplesizebutduringthedatacollectionphaseyou wait to reach a point of data saturation. When you are not getting new information or it isnegligible, it is assumedyou have reached a data saturation point and you stop collecting additionalinformation.Considerable importance isplacedon thesamplesize inquantitative research,dependingupon the
typeofstudyandthepossibleuseofthefindings.Studieswhicharedesignedtoformulatepolicies,totestassociationsorrelationships,or toestablishimpactassessmentsplaceaconsiderableemphasisonlargesamplesize.Thisisbasedupontheprinciplethatalargersamplesizewillensuretheinclusionofpeoplewithdiversebackgrounds, thusmakingthesamplerepresentativeof thestudypopulation.Thesamplesizeinqualitativeresearchdoesnotplayanysignificantroleasthepurposeistostudyonlyoneorafewcasesinordertoidentifythespreadofdiversityandnotitsmagnitude.Insuchsituationsthedatasaturationstageduringdatacollectiondeterminesthesamplesize.In quantitative research, randomisation is used to avoid bias in the selection of a sample and is
selectedinsuchawaythatitrepresentsthestudypopulation.Inqualitativeresearchnosuchattemptismadeinselectingasample.Youpurposelyselect‘information-rich’respondentswhowillprovideyouwiththeinformationyouneed.Inquantitativeresearch,thisisconsideredabiasedsample.Mostof thesamplingstrategies, includingsomenon-probabilityones,describedinthischaptercan
be used when undertaking a quantitative study provided it meets the requirements. However, whenconductingaqualitativestudyonlythenon-probabilitysamplingdesignscanbeused.
FIGURE12.1Theconceptofsampling
Samplinginquantitativeresearch
Theconceptofsampling
Letustakeaverysimpleexampletoexplaintheconceptofsampling.Supposeyouwanttoestimatetheaverage age of the students in your class. There are twoways of doing this. The firstmethod is tocontactallstudentsintheclass,findouttheirages,addthemupandthendividethisbythenumberofstudents(theprocedureforcalculatinganaverage).Thesecondmethodistoselectafewstudentsfromtheclass,askthemtheirages,addthemupandthendividebythenumberofstudentsyouhaveasked.Fromthisyoucanmakeanestimateoftheaverageageoftheclass.Similarly,supposeyouwanttofindouttheaverageincomeoffamilieslivinginacity.Imaginetheamountofeffortandresourcesrequiredtogo toevery family in thecity to findout their income!Youcould instead select a few families tobecomethebasisofyourenquiryandthen,fromwhatyouhavefoundoutfromthefewfamilies,makeanestimateoftheaverageincomeoffamiliesinthecity.Similarly,electionopinionpollscanbeused.Thesearebaseduponaverysmallgroupofpeoplewhoarequestionedabouttheirvotingpreferencesand,onthebasisoftheseresults,apredictionismadeabouttheprobableoutcomeofanelection.Sampling,therefore,istheprocessofselectingafew(asample)fromabiggergroup(thesampling
population) tobecome thebasis for estimatingorpredicting theprevalenceof anunknownpieceofinformation,situationoroutcomeregardingthebiggergroup.Asampleisasubgroupofthepopulationyouareinterestedin.SeeFigure12.1.Thisprocessofselectingasamplefromthetotalpopulationhasadvantagesanddisadvantages.The
advantagesarethatitsavestimeaswellasfinancialandhumanresources.However,thedisadvantageisthatyoudonotfindouttheinformationaboutthepopulation’scharacteristicsofinteresttoyoubutonlyestimateorpredictthem.Hence,thepossibilityofanerrorinyourestimationexists.Sampling,therefore,isatrade-offbetweencertainbenefitsanddisadvantages.Whileontheonehand
you save time and resources, on the other hand youmay compromise the level of accuracy in yourfindings.Throughsamplingyouonlymakeanestimateabouttheactualsituationprevalentinthetotalpopulation from which the sample is drawn. If you ascertain a piece of information from the totalsampling population, and if your method of enquiry is correct, your findings should be reasonablyaccurate.However,ifyouselectasampleandusethisasthebasisfromwhichtoestimatethesituationin the total population, an error is possible. Tolerance of this possibility of error is an importantconsiderationinselectingasample.
Samplingterminology
Letus,again,considertheexamplesusedabovewhereourmainaimsaretofindouttheaverageageofthe class, the average incomeof the families living in the city and the likely election outcome for aparticularstateorcountry.Letusassumethatyouadoptthesamplingmethod–thatis,youselectafewstudents,familiesorelectoratestoachievetheseaims.Inthisprocessthereareanumberofaspects:
Theclass,familieslivinginthecityorelectoratesfromwhichyouselectyouselectyoursamplearecalledthepopulationorstudypopulation,andareusuallydenotedbytheletterN.Thesmallgroupofstudents,familiesorelectorsfromwhomyoucollecttherequiredinformationtoestimatetheaverageageoftheclass,averageincomeortheelectionoutcomeiscalledthesample.
Thenumberofstudents,familiesorelectorsfromwhomyouobtaintherequiredinformationiscalledthesamplesizeandisusuallydenotedbythelettern.Thewayyouselectstudents,familiesorelectorsiscalledthesamplingdesignorsamplingstrategy.Eachstudent,familyorelectorthatbecomesthebasisforselectingyoursampleiscalledthesamplingunitorsamplingelement.Alistidentifyingeachstudent,familyorelectorinthestudypopulationiscalledthesamplingframe.Ifallelementsinasamplingpopulationcannotbeindividuallyidentified,youcannothaveasamplingframeforthatstudypopulation.Yourfindingsbasedontheinformationobtainedfromyourrespondents(sample)arecalledsamplestatistics.Yoursamplestatisticsbecomethebasisofestimatingtheprevalenceoftheabovecharacteristicsinthestudypopulation.Yourmainaimistofindanswerstoyourresearchquestionsinthestudypopulation,notinthesampleyoucollectedinformationfrom.Fromsamplestatisticswemakeanestimateoftheanswerstoourresearchquestionsinthestudypopulation.Theestimatesarrivedatfromsamplestatisticsarecalledpopulationparametersorthepopulationmean.
Principlesofsampling
The theory of sampling is guided by three principles. To effectively explain these, we will take anextremelysimpleexample.SupposetherearefourindividualsA,B,CandD.FurthersupposethatAis18yearsofage,Bis20,Cis23andDis25.Asyouknowtheirages,youcanfindout(calculate)theiraverageagebysimplyadding18+20+23+25=86anddividingby4.Thisgivestheaverage(mean)ageofA,B,CandDas21.5years.Nowletussupposethatyouwanttoselectasampleoftwoindividualstomakeanestimateofthe
averageageofthefourindividuals.Toselectanunbiasedsample,weneedtomakesurethateachunithas an equal and independent chance of selection in the sample.Randomisation is a process thatenables you to achieve this. In order to achieve randomisation we use the theory of probability informingpairswhichwillprovideuswithsixpossiblecombinationsoftwo:AandB;AandC;AandD;BandC;BandD; andCandD.Letus takeeachof thesepairs to calculate the averageageof thesample:
1. A+B=18+20=38/2=19.0years;2. A+C=18+23=41/2=20.5years;3. A+D=18+25=43/2=21.5years;4. B+C=20+23=43/2=21.5years;5. B+D=20+25=45/2=22.5years;6. C+D=23+25=48/2=24.0years.
Notice that inmost cases theaverageagecalculatedon thebasisof these samplesof two (samplestatistics)isdifferent.Nowcomparethesesamplestatisticswiththeaverageofallfourindividuals–thepopulation mean (population parameter) of 21.5 years. Out of a total of six possible samplecombinations, only in the case of two is there no difference between the sample statistics and thepopulationmean.Wherethereisadifference,thisisattributedtothesampleandisknownassamplingerror. Again, the size of the sampling error varies markedly. Let us consider the difference in thesamplestatisticsandthepopulationmeanforeachofthesixsamples(Table12.1).
TABLE12.1Thedifferencebetweensamplestatisticsandthepopulationmean
Thisanalysissuggestsaveryimportantprincipleofsampling:
Principle1 – in amajority of cases of sampling therewill be a difference between the samplestatisticsand the truepopulationmean,which is attributable to the selectionof theunits in thesample.
Tounderstandthesecondprinciple,letuscontinuewiththeaboveexample,butinsteadofasampleoftwoindividualswetakeasampleofthree.Therearefourpossiblecombinationsofthreethatcanbedrawn:
1. 1A+B+C=18+20+23=61/3=20.33years;2. 2A+B+D=18+20+25=63/3=21.00years;3. 3A+C+D=18+23+25=66/3=22.00years;4. 4B+C+D=20+23+25=68/3=22.67years.
Now, let us compare the difference between the sample statistics and the populationmean (Table12.2).
TABLE12.2Thedifferencebetweenasampleandapopulationaverage
Compare the differences calculated in Table 12.1 and Table 12.2. In Table 12.1 the differencebetweenthesamplestatisticsandthepopulationmeanliesbetween–2.5and+2.5years,whereasintheseconditisbetween–1.17and+1.17years.Thegapbetweenthesamplestatisticsandthepopulationmean is reduced in Table 12.2. This reduction is attributed to the increase in the sample size. This,therefore,leadstothesecondprinciple:
Principle2– thegreater the sample size, themoreaccurate the estimateof the truepopulationmean.
Thethirdprincipleofsamplingisparticularlyimportantasanumberofsamplingstrategies,suchasstratified andcluster sampling, arebasedon it.Tounderstand thisprinciple, let us continuewith thesame example but use slightly different data. Suppose the ages of four individuals are markedly
different:A=18,B=26,C=32andD=40.Inotherwords,wearevisualisingapopulationwheretheindividualswithrespecttoage–thevariableweareinterestedin–aremarkedlydifferent.Letusfollowthesameprocedure,selectingsamplesoftwoindividualsatatimeandthenthree.Ifwe
workthroughthesameprocedures(describedabove)wewillfindthatthedifferenceintheaverageageinthecaseofsamplesoftworangesbetween–7.00and+7.00yearsandinthecaseofthesampleofthree ranges between –3.67 and +3.67. In both cases the range of the difference is greater thanpreviouslycalculated.Thisisattributabletothegreaterdifferenceintheagesofthefourindividuals–the sampling population. In other words, the sampling population is more heterogeneous (varied ordiverse)inregardtoage.
Principle3 – the greater the difference in the variable under study in a population for a givensamplesize,thegreaterthedifferencebetweenthesamplestatisticsandthetruepopulationmean.
Theseprinciplesarecrucialtokeepinmindwhenyouaredeterminingthesamplesizeneededforaparticularlevelofaccuracy,andinselectingthesamplingstrategybestsuitedtoyourstudy.
Factorsaffectingtheinferencesdrawnfromasample
Theaboveprinciplessuggestthattwofactorsmayinfluencethedegreeofcertaintyabouttheinferencesdrawnfromasample:
1. Thesizeofthesample–Findingsbaseduponlargersampleshavemorecertaintythanthosebasedonsmallerones.Asarule,thelargerthesamplesize,themoreaccuratethefindings.
2. Theextentofvariationinthesamplingpopulation–Thegreaterthevariationinthestudypopulationwithrespecttothecharacteristicsunderstudy,foragivensamplesize,thegreatertheuncertainty.(Intechnicalterms,thegreaterthestandarddeviation,thehigherthestandarderrorforagivensamplesizeinyourestimates.)Ifapopulationishomogeneous(uniformorsimilar)withrespecttothecharacteristicsunderstudy,asmallsamplecanprovideareasonablygoodestimate,butifitisheterogeneous(dissimilarordiversified),youneedtoselectalargersampletoobtainthesamelevelofaccuracy.Ofcourse,ifalltheelementsinapopulationareidentical,thentheselectionofevenonewillprovideanabsolutelyaccurateestimate.Asarule,thehigherthevariationwithrespecttothecharacteristicsunderstudyinthestudypopulation,thegreatertheuncertaintyforagivensamplesize.
Aimsinselectingasample
When you select a sample in quantitative studies you are primarily aiming to achieve maximumprecisioninyourestimateswithinagivensamplesize,andavoidbiasintheselectionofyoursample.Biasintheselectionofasamplecanoccurif:
samplingisdonebyanon-randommethod–thatis,iftheselectionisconsciouslyorunconsciouslyinfluencedbyhumanchoice;thesamplingframe–list,indexorotherpopulationrecords–whichservesasthebasisof
selection,doesnotcoverthesamplingpopulationaccuratelyandcompletely;asectionofasamplingpopulationisimpossibletofindorrefusestoco-operate.
Typesofsampling
Thevarioussamplingstrategiesinquantitativeresearchcanbecategorisedasfollows(Figure12.2):
FIGURE12.2Typesofsamplinginquantitativeresearch
random/probabilitysamplingdesigns;non-random/non-probabilitysamplingdesignsselectingapredeterminedsamplesize;‘mixed’samplingdesign.
Tounderstandthesedesigns,wewilldiscusseachtypeindividually.
Random/probabilitysamplingdesigns
Foradesigntobecalledrandomsamplingorprobabilitysampling,itisimperativethateachelementinthepopulationhasanequalandindependentchanceofselectioninthesample.Equalimpliesthatthe
probabilityofselectionofeachelementinthepopulationisthesame;thatis,thechoiceofanelementin the sample is not influenced by other considerations such as personal preference. The concept ofindependencemeansthatthechoiceofoneelementisnotdependentuponthechoiceofanotherelementin the sampling; that is, the selection or rejection of one element does not affect the inclusion orexclusionofanother.Toexplaintheseconceptsletusreturntoourexampleoftheclass.Supposethereare80studentsintheclass.Assume20oftheserefusetoparticipateinyourstudy.You
wanttheentirepopulationof80studentsinyourstudybut,as20refusetoparticipate,youcanonlyuseasampleof60students.The20studentswhorefusetoparticipatecouldhavestrongfeelingsabouttheissuesyouwishtoexplore,butyourfindingswillnotreflecttheiropinions.Theirexclusionfromyourstudymeans thateachof the80studentsdoesnothaveanequalchanceofselection.Therefore,yoursampledoesnotrepresentthetotalclass.Thesamecouldapplytoacommunity.Inacommunity,inadditiontotherefusaltoparticipate,letus
assume that you are unable to identify all the residents living in the community. If a significantproportion of people cannot be included in the sampling population because they either cannot beidentifiedor refuse toparticipate, thenanysampledrawnwillnotgiveeachelement in thesamplingpopulationanequalchanceofbeingselectedinthesample.Hence,thesamplewillnotberepresentativeofthetotalcommunity.Tounderstand theconceptof an independent chanceof selection, letusassume that thereare five
studentsintheclasswhoareextremelyclosefriends.Ifoneofthemisselectedbutrefusestoparticipatebecausetheotherfourarenotchosen,andyouarethereforeforcedtoselecteitherthefiveornone,thenyoursamplewillnotbeconsideredanindependentsamplesincetheselectionofoneisdependentupontheselectionofothers.Thesamecouldhappeninthecommunitywhereasmallgroupsaysthateitherallofthemornoneofthemwillparticipateinthestudy.Inthesesituationswhereyouareforcedeitherto include or to exclude a part of the sampling population, the sample is not considered to beindependent, and hence is not representative of the sampling population.However, if the number ofrefusalsisfairlysmall,inpracticalterms,itshouldnotmakethesamplenon-representative.Inpracticetherearealwayssomepeoplewhodonotwanttoparticipateinthestudybutyouonlyneedtoworryifthenumberissignificantlylarge.A sample canonlybe considereda random/probability sample (and therefore representativeof the
population under study) if both these conditions aremet.Otherwise, bias can be introduced into thestudy.Therearetwomainadvantagesofrandom/probabilitysamples:
1. Astheyrepresentthetotalsamplingpopulation,theinferencesdrawnfromsuchsamplescanbegeneralisedtothetotalsamplingpopulation.
2. Somestatisticaltestsbaseduponthetheoryofprobabilitycanbeappliedonlytodatacollectedfromrandomsamples.Someofthesetestsareimportantforestablishingconclusivecorrelations.
Methodsofdrawingarandomsample
Ofthemethodsthatyoucanadopttoselectarandomsamplethethreemostcommonare:
1. Thefishbowldraw–ifyourtotalpopulationissmall,aneasyprocedureistonumbereachelementusingseparateslipsofpaperforeachelement,putalltheslipsintoaboxandthenpickthemoutonebyonewithoutlooking,untilthenumberofslipsselectedequalsthesamplesizeyou
decidedupon.Thismethodisusedinsomelotteries.2. Computerprogram–thereareanumberofprogramsthatcanhelpyoutoselectarandom
sample.3. Atableofrandomlygeneratednumbers–mostbooksonresearchmethodologyandstatistics
includeatableofrandomlygeneratednumbersintheirappendices(see,e.g.,Table12.3).YoucanselectyoursampleusingthesetablesaccordingtotheproceduredescribedinFigure12.3.
Theprocedureforselectingasampleusingatableofrandomnumbersisasfollows:Letus takeanexample to illustrate theuseofTable12.3for randomnumbers.Letusassumethat
yoursamplingpopulationconsistsof256individuals.Numbereachindividualfrom1to256.Randomlyselectthestartingpage,setofcolumn(1to10)orrowfromthetableandthenidentifythreecolumnsorrowsofnumbers.Suppose you identify the ninth column of numbers and the last three digits of this column
(underlined). Assume that you are selecting 10 per cent of the total population as your sample (25elements).Letusgothroughthenumbersunderlinedin theninthsetofcolumns.Thefirstnumber is049whichisbelow256(totalpopulation);hence,the49thelementbecomesapartofyoursample.Thesecondnumber,319,ismorethanthetotalelementsinyourpopulation(256);hence,youcannotacceptthe319thelementinthesample.Thesameappliestothenextelement,758,andindeedthenextfiveelements, 589, 507, 483, 487 and 540.After 540 is 232, and as this number iswithin the samplingframe,itcanbeacceptedasapartofthesample.Similarly,ifyoufollowdownthesamethreedigitsinthesamecolumn,youselect052,029,065,246and161,beforeyoucometotheelement029again.Asthe29thelementhasalreadybeenselected,go to thenextnumber,andsoonuntil25elementshavebeen chosen. Once you have reached the end of a column, you can either move to the next set ofcolumnsorrandomlyselectanotheroneinordertocontinuetheprocessofselection.Forexample,the25elementsshowninTable12.4areselectedfromtheninth,tenthandsecondcolumnsofTable12.3.
TABLE12.3Selectingasampleusingatableforrandomnumbers
Source:StatisticalTables,3e,byF.JamesRohlfandRobertR.Sokal.Copyright©1969,1981,1994byW.H.FreemanandCompany.Usedwithpermission.
FIGURE12.3Theprocedureforusingatableofrandomnumbers
TABLE12.4Selectedelementsusingthetableofrandomnumbers
Samplingwithorwithoutreplacement
Randomsamplingcanbeselectedusingtwodifferentsystems:
1. samplingwithoutreplacement;2. samplingwithreplacement.
Supposeyouwanttoselectasampleof20studentsoutofatotalof80.Thefirststudentisselectedoutofthetotalclass,andsotheprobabilityofselectionfor thefirststudent is1/80.Whenyouselect thesecondstudentthereareonly79leftintheclassandtheprobabilityofselectionforthesecondstudentisnot1/80but1/79.Theprobabilityofselecting thenext student is1/78.By the timeyouselect the20th student, the probability of his/her selection is 1/61. This type of sampling is called samplingwithoutreplacement.Butthisiscontrarytoourbasicdefinitionofrandomisation;thatis,eachelementhas an equal and independent chance of selection. In the second system, called sampling withreplacement,theselectedelementisreplacedinthesamplingpopulationandifitisselectedagain,itisdiscarded and the next one is selected. If the sampling population is fairly large, the probability ofselectingthesameelementtwiceisfairlyremote.
FIGURE12.4Theprocedureforselectingasimplerandomsample
Specificrandom/probabilitysamplingdesigns
Therearethreecommonlyusedtypesofrandomsamplingdesign.
1. Simplerandomsampling(SRS)–Themostcommonlyusedmethodofselectingaprobabilitysample.Inlinewiththedefinitionofrandomisation,wherebyeachelementinthepopulationisgivenanequalandindependentchanceofselection,asimplerandomsampleisselectedbytheprocedurepresentedinFigure12.4.Toillustrate,letusagaintakeourexampleoftheclass.Thereare80studentsintheclass,andsothefirststepistoidentifyeachstudentbyanumberfrom1to80.Supposeyoudecidetoselectasampleof20usingthesimplerandomsamplingtechnique.Usethefishbowldraw,thetableforrandomnumbersoracomputerprogramtoselectthe20students.These20studentsbecomethebasisofyourenquiry.
2. Stratifiedrandomsampling–Asdiscussed,theaccuracyofyourestimatelargelydependsontheextentofvariabilityorheterogeneityofthestudypopulationwithrespecttothecharacteristicsthathaveastrongcorrelationwithwhatyouaretryingtoascertain(Principle3).Itfollows,therefore,thatiftheheterogeneityinthepopulationcanbereducedbysomemeansforagivensamplesizeyoucanachievegreateraccuracyinyourestimate.Stratifiedrandomsamplingisbaseduponthislogic.Instratifiedrandomsamplingtheresearcherattemptstostratifythepopulationinsuchawaythatthepopulationwithinastratumishomogeneouswithrespecttothecharacteristiconthebasisofwhichitisbeingstratified.Itisimportantthatthecharacteristicschosenasthebasisof
stratificationareclearlyidentifiableinthestudypopulation.Forexample,itismucheasiertostratifyapopulationonthebasisofgenderthanonthebasisofage,incomeorattitude.Itisalsoimportantforthecharacteristicthatbecomesthebasisofstratificationtoberelatedtothemainvariablethatyouareexploring.Oncethesamplingpopulationhasbeenseparatedintonon-overlappinggroups,youselecttherequirednumberofelementsfromeachstratum,usingthesimplerandomsamplingtechnique.Therearetwotypesofstratifiedsampling:proportionatestratifiedsamplinganddisproportionatestratifiedsampling.Withproportionatestratifiedsampling,thenumberofelementsfromeachstratuminrelationtoitsproportioninthetotalpopulationisselected,whereasindisproportionatestratifiedsampling,considerationisnotgiventothesizeofthestratum.TheprocedureforselectingastratifiedsampleisschematicallypresentedinFigure12.5.
3. Clustersampling–Simplerandomandstratifiedsamplingtechniquesarebasedonaresearcher’sabilitytoidentifyeachelementinapopulation.Itiseasytodothisifthetotalsamplingpopulationissmall,butifthepopulationislarge,asinthecaseofacity,stateorcountry,itbecomesdifficultandexpensivetoidentifyeachsamplingunit.Insuchcasestheuseofclustersamplingismoreappropriate.Clustersamplingisbasedontheabilityoftheresearchertodividethesamplingpopulationintogroups(baseduponvisibleoreasilyidentifiablecharacteristics),calledclusters,andthentoselectelementswithineachcluster,usingtheSRStechnique.Clusterscanbeformedonthebasisofgeographicalproximityoracommoncharacteristicthathasacorrelationwiththemainvariableofthestudy(asinstratifiedsampling).Dependingonthelevelofclustering,sometimessamplingmaybedoneatdifferentlevels.Theselevelsconstitutethedifferentstages(single,doubleormultiple)ofclustering,whichwillbeexplainedlater.Imagineyouwanttoinvestigatetheattitudeofpost-secondarystudentsinAustraliatowardsproblemsinhighereducationinthecountry.HighereducationinstitutionsareineverystateandterritoryofAustralia.Inaddition,therearedifferenttypesofinstitutions,forexampleuniversities,universitiesoftechnology,collegesofadvancededucationandcollegesoftechnicalandfurthereducation(TAFE)(Figure12.6).Withineachinstitutionvariouscoursesareofferedatbothundergraduateandpostgraduatelevels.Eachacademiccoursecouldtakethreetofouryears.Youcanimaginethemagnitudeofthetask.Insuchsituationsclustersamplingisextremelyusefulinselectingarandomsample.Thefirstlevelofclustersamplingcouldbeatthestateorterritorylevel.Clusterscouldbegroupedaccordingtosimilarcharacteristicsthatensuretheircomparabilityintermsofstudentpopulation.Ifthisisnoteasy,youmaydecidetoselectallthestatesandterritoriesandthenselectasampleattheinstitutionallevel.Forexample,withasimplerandomtechnique,oneinstitutionfromeachcategorywithineachstatecouldbeselected(oneuniversity,oneuniversityoftechnologyandoneTAFEcollege).Thisisbasedupontheassumptionthatinstitutionswithinacategoryarefairlysimilarwithregardstostudentprofile.Then,withinaninstitutiononarandombasis,oneormoreacademicprogrammescouldbeselected,dependingonresources.Withineachstudyprogrammeselected,studentsstudyinginaparticularyearcouldthenbeselected.Further,selectionofaproportionofstudentsstudyinginaparticularyearcouldthenbemadeusingtheSRStechnique.Theprocessofselectingasampleinthismanneriscalledmulti-stageclustersampling.
FIGURE12.5Theprocedureforselectingastratifiedsample
FIGURE12.6Theconceptofclustersampling
Non-random/non-probabilitysamplingdesignsinquantitativeresearch
Non-probabilitysamplingdesignsdonot follow the theoryofprobability in thechoiceofelementsfromthesamplingpopulation.Non-probabilitysamplingdesignsareusedwhenthenumberofelementsinapopulationiseitherunknownorcannotbeindividuallyidentified.Insuchsituationstheselectionofelementsisdependentuponotherconsiderations.Therearefivecommonlyusednon-randomdesigns,eachbasedonadifferentconsideration,whicharecommonlyusedinbothqualitativeandquantitativeresearch.Theseare:
1. quotasampling;2. accidentalsampling;3. judgementalsamplingorpurposivesampling;4. expertsampling;5. snowballsampling.
Whatdifferentiatesthesedesignsbeingtreatedasquantitativeorqualitativeisthepredeterminedsamplesize.Inquantitativeresearchyouusethesedesignstoselectapredeterminednumberofcases(samplesize), whereas in qualitative research you do not decide the number of respondents in advance butcontinue to select additional cases till you reach the data saturation point. In addition, in qualitativeresearch, you will predominantly use judgemental and accidental sampling strategies to select yourrespondents.Expertsamplingisverysimilartojudgementalsamplingexceptthatinexpertsamplingthesampling population comprises experts in the field of enquiry.You can also use quota and snowballsamplinginqualitativeresearchbutwithouthavingapredeterminednumberofcasesinmind(samplesize).
Quotasampling
The main consideration directing quota sampling is the researcher’s ease of access to the samplepopulation.Inadditiontoconvenience,youareguidedbysomevisiblecharacteristic,suchasgenderorrace,ofthestudypopulationthatisofinteresttoyou.Thesampleisselectedfromalocationconvenienttoyouasaresearcher,andwheneverapersonwiththisvisiblerelevantcharacteristicisseenthatpersonis asked to participate in the study. The process continues until you have been able to contact therequirednumberofrespondents(quota).Letussupposethatyouwanttoselectasampleof20malestudentsinordertofindouttheaverage
ageofthemalestudentsinyourclass.Youdecidetostandattheentrancetotheclassroom,asthisisconvenient,andwheneveramalestudententerstheclassroom,youaskhisage.Thisprocesscontinuesuntilyouhaveasked20studentstheirage.Alternatively,youmightwanttofindoutabouttheattitudesof Aboriginal and Torres Strait Islander students towards the facilities provided to them in youruniversity.Youmightstandataconvenientlocationand,wheneveryouseesuchastudent,collecttherequiredinformationthroughwhatevermethodofdatacollection(suchasinterviewing,questionnaire)youhaveadoptedforthestudy.Theadvantagesofusingthisdesignare:itistheleastexpensivewayofselectingasample;youdo
not need any information, such as a sampling frame, the total numberof elements, their location, orotherinformationaboutthesamplingpopulation;anditguaranteestheinclusionofthetypeofpeopleyouneed.Thedisadvantagesare:astheresultingsampleisnotaprobabilityone,thefindingscannotbegeneralised to the total sampling population; and the most accessible individuals might havecharacteristicsthatareuniquetothemandhencemightnotbetrulyrepresentativeofthetotalsamplingpopulation.You canmake your samplemore representative of your study population by selecting itfromvariouslocationswherepeopleofinteresttoyouarelikelytobeavailable.
Accidentalsampling
Accidental sampling is also based upon convenience in accessing the sampling population.Whereasquota sampling attempts to include people possessing an obvious/visible characteristic, accidental
samplingmakes no such attempt. You stop collecting data when you reach the required number ofrespondentsyoudecidedtohaveinyoursample.Thismethodofsamplingiscommonamongmarketresearchandnewspaperreporters.Ithasmoreor
lessthesameadvantagesanddisadvantagesasquotasamplingbut,inaddition,asyouarenotguidedbyanyobviouscharacteristics,somepeoplecontactedmaynothavetherequiredinformation.
Judgementalorpurposivesampling
The primary consideration in purposive sampling is your judgement as towho can provide the bestinformationtoachievetheobjectivesofyourstudy.Youasaresearcheronlygotothosepeoplewhoinyouropinionarelikelytohavetherequiredinformationandbewillingtoshareitwithyou.Thistypeofsamplingisextremelyusefulwhenyouwanttoconstructahistoricalreality,describea
phenomenonordevelopsomethingaboutwhichonlyalittleisknown.Thissamplingstrategyismorecommoninqualitativeresearch,butwhenyouuseitinquantitativeresearchyouselectapredeterminednumberofpeoplewho,inyourjudgement,arebestpositionedtoprovideyoutheneededinformationforyourstudy.
Expertsampling
Theonlydifferencebetweenjudgementalsamplingandexpertsamplingisthatinthecaseoftheformerit isentirelyyour judgementas to theabilityof therespondents tocontribute to thestudy.But in thecaseofexpertsampling,yourrespondentsmustbeknownexpertsinthefieldofinteresttoyou.Thisisagain used in both types of research butmore so in qualitative research studies.Whenyou use it inqualitative research, the number of people you talk to is dependent upon the data saturation pointwhereas in quantitative research you decide on the number of experts to be contacted withoutconsideringthesaturationpoint.Youfirst identifypersonswithdemonstratedorknownexpertise inanareaof interest toyou,seek
theirconsentforparticipation,andthencollecttheinformationeitherindividuallyorcollectivelyintheformofagroup.
FIGURE12.7Snowballsampling
Snowballsampling
Snowballsamplingistheprocessofselectingasampleusingnetworks.Tostartwith,afewindividualsinagroupororganisationareselectedand therequired information iscollectedfromthem.Theyarethen asked to identify other people in the group or organisation, and the people selected by thembecomeapartof the sample. Information iscollected from them,and then thesepeopleareasked toidentify other members of the group and, in turn, those identified become the basis of further datacollection(Figure12.7).Thisprocessiscontinueduntiltherequirednumberorasaturationpointhasbeenreached,intermsoftheinformationbeingsought.This sampling technique is useful if you know little about the group or organisation youwish to
study,asyouneedonly tomakecontactwithafewindividuals,whocanthendirectyouto theothermembers of the group. This method of selecting a sample is useful for studying communicationpatterns, decisionmaking or diffusion of knowledgewithin a group. There are disadvantages to thistechnique, however. The choice of the entire sample rests upon the choice of individuals at the firststage.If theybelongtoaparticularfactionorhavestrongbiases, thestudymaybebiased.Also, it isdifficulttousethistechniquewhenthesamplebecomesfairlylarge.
Systematicsamplingdesign:a‘mixed’design
Systematicsamplinghasbeenclassifiedasa‘mixed’samplingdesignbecauseithasthecharacteristicsofbothrandomandnon-randomsamplingdesigns.Insystematicsamplingthesamplingframeisfirstdividedintoanumberofsegmentscalledintervals.
Then, from the first interval, using the SRS technique, one element is selected. The selection ofsubsequentelementsfromotherintervalsisdependentupontheorderoftheelementselectedinthefirstinterval.Ifinthefirstintervalitisthefifthelement,thefifthelementofeachsubsequentintervalwillbechosen.Noticethatfromthefirstintervalthechoiceofanelementisonarandombasis,butthechoiceoftheelementsfromsubsequentintervalsisdependentuponthechoicefromthefirst,andhencecannotbe classified as a random sample.The procedure used in systematic sampling is presented inFigure12.8.
FIGURE12.8TheprocedureforselectingasystematicsampleAlthough the general procedure for selecting a sample by the systematic sampling technique is
describedabove,youcandeviate fromitbyselectingadifferentelement fromeach intervalwith theSRS technique. By adopting this, systematic sampling can be classified under probability samplingdesigns.To select a random sample you must have a sampling frame (Figure 12.9). Sometimes this is
impossible,orobtainingonemaybe tooexpensive.However, inreal life therearesituationswhereakindofsamplingframeexists,forexamplerecordsofclientsinanagency,enrolmentlistsofstudentsinaschooloruniversity,electorallistsofpeoplelivinginanarea,orrecordsofthestaffemployedinanorganisation.Allthesecanbeusedasasamplingframetoselectasamplewiththesystematicsamplingtechnique.Thisconvenienceofhavinga‘ready-made’samplingframemaybeataprice:insomecasesitmaynottrulybearandomlisting.Mostlytheselistsareinalphabeticalorder,baseduponanumber
assignedtoacase,orarrangedinawaythatisconvenienttotheusersoftherecords.Ifthe‘widthofaninterval’ is large, say, 1 in 30 cases, and if the cases are arranged in alphabetical order, you couldprecludesomewhosesurnamesstartwiththesameletterorsomeadjoininglettermaynotbeincludedatall.Suppose there are 50 students in a class and youwant to select 10 students using the systematic
samplingtechnique.Thefirststepistodeterminethewidthoftheinterval(50/10=5).Thismeansthatfromeveryfiveyouneedtoselectoneelement.UsingtheSRStechnique,fromthefirstinterval(1–5elements), select one of the elements. Suppose you selected the third element. From the rest of theintervalsyouwouldselecteverythirdelement.
Thecalculationofsamplesize
Studentsandothersoftenask:‘HowbigasampleshouldIselect?’,‘Whatshouldbemysamplesize?’and‘HowmanycasesdoIneed?’Basically,itdependsonwhatyouwanttodowiththefindingsandwhat type of relationships you want to establish. Your purpose in undertaking research is the maindeterminantofthelevelofaccuracyrequiredintheresults,andthislevelofaccuracyisanimportantdeterminantofsamplesize.However,inqualitativeresearch,asthemainfocusistoexploreordescribea situation, issue,processorphenomenon, thequestionof sample size is less important.Youusuallycollectdata tillyouthinkyouhavereachedsaturationpoint in termsofdiscoveringnewinformation.Onceyouthinkyouarenotgettingmuchnewdatafromyourrespondents,youstopcollectingfurtherinformation.Ofcourse,thediversityorheterogeneityinwhatyouaretryingtofindoutaboutplaysanimportantroleinhowfastyouwillreachsaturationpoint.Andremember:thegreatertheheterogeneityordiversityinwhatyouaretryingtofindoutabout,thegreaterthenumberofrespondentsyouneedtocontacttoreachsaturationpoint.Indeterminingthesizeofyoursampleforquantitativestudiesandinparticularforcause-and-effectstudies,youneedtoconsiderthefollowing:
FIGURE12.9Systematicsampling
Atwhatlevelofconfidencedoyouwanttotestyourresults,findingsorhypotheses?Withwhatdegreeofaccuracydoyouwishtoestimatethepopulationparameters?Whatistheestimatedlevelofvariation(standarddeviation),withrespecttothemainvariableyou
arestudying,inthestudypopulation?
Answeringthesequestionsisnecessaryregardlessofwhetheryouintendtodeterminethesamplesizeyourselforhaveanexpertdoitforyou.Thesizeofthesampleisimportantfortestingahypothesisorestablishinganassociation,butforotherstudiesthegeneralruleis:thelargerthesamplesize,themoreaccurate your estimates. In practice, your budget determines the size of your sample.Your skills inselectingasample,withintheconstraintsofyourbudget,lieinthewayyouselectyourelementssothattheyeffectivelyandadequatelyrepresentyoursamplingpopulation.To illustrate this procedure let us take the example of a class. Suppose you want to find out the
averageageofthestudentswithinanaccuracyof0.5ofayear;thatis,youcantolerateanerrorofhalfayearoneithersideof the trueaverageage.Letusalsoassumethatyouwant to find theaverageagewithinhalfayearofaccuracyat the95percentconfidencelevel; that is,youwanttobe95percentconfidentaboutyourfindings.Theformula(fromstatistics)fordeterminingtheconfidencelimitsis
where=estimatedvalueofthepopulationmean=averageagecalculatedfromthesamplet0.05=valueoftat95percentconfidencelevel
σ/√η=standarderrorσ=standarddeviationη=samplesize√=squarerootIfwedecidetotolerateanerrorofhalfayear,thatmeans
*t-valuefromthefollowingtable
Thereisonlyoneunknownquantityintheaboveequation,thatisσ.Now themainproblem is to find thevalueofσwithouthaving tocollectdata.This is thebiggest
probleminestimatingthesamplesize.Becauseofthisitisimportanttoknowasmuchaspossibleaboutthestudypopulation.Thevalueofσcanbefoundbyoneofthefollowing:
1. guessing;2. consultinganexpert;3. obtainingthevalueofσfrompreviouscomparablestudies;or4. carryingoutapilotstudytocalculatethevalue.
Letusassumethatσis1year.Then
Hence,todeterminetheaverageageoftheclassatalevelof95percentaccuracy(assumingσ=1year)withhalfayearoferror,asampleofatleast16studentsisnecessary.Nowassumethat, insteadof95percent,youwant tobe99percentconfidentabout theestimated
age,toleratinganerrorofhalfayear.Then
Hence,ifyouwanttobe99percentconfidentandarewillingtotolerateanerrorofhalfayear,youneedtoselectasampleof27students.Similarly,youcancalculatethesamplesizewithvaryingvaluesofσ.Rememberthegoldenrule:thegreateristhesamplesize,themoreaccuratelyyourfindingswillreflectthe‘true’picture.
Samplinginqualitativeresearch
Asthemainaiminqualitativeenquiriesistoexplorethediversity,samplesizeandsamplingstrategydonotplayasignificantroleintheselectionofasample.Ifselectedcarefully,diversitycanbeextensivelyand accurately described on the basis of information obtained even from one individual. All non-probabilitysamplingdesigns–purposive, judgemental,expert,accidentalandsnowball–canalsobeusedinqualitativeresearchwithtwodifferences:
1. Inquantitativestudiesyoucollectinformationfromapredeterminednumberofpeoplebut,inqualitativeresearch,youdonothaveasamplesizeinmind.Datacollectionbaseduponapredeterminedsamplesizeandthesaturationpointdistinguishestheiruseinquantitativeandqualitativeresearch.
2. Inquantitativeresearchyouareguidedbyyourdesiretoselectarandomsample,whereasinqualitativeresearchyouareguidedbyyourjudgementastowhoislikelytoprovideyouwiththe‘best’information.
Theconceptofsaturationpointinqualitativeresearch
As you already know, in qualitative research data is usually collected to a point where you are notgettingnewinformationoritisnegligible–thedatasaturationpoint.Thisstagedeterminesthesamplesize.Itisimportantforyoutokeepinmindthattheconceptofdatasaturationpointishighlysubjective.It
isyouwhoarecollectingthedataanddecidewhenyouhaveattainedthesaturationpointinyourdatacollection. How soon you reach the saturation point depends upon how diverse is the situation orphenomenon that you are studying.The greater the diversity, the greater the number of people fromwhomyouneedtocollecttheinformationtoreachthesaturationpoint.Theconceptofsaturationpointismoreapplicabletosituationswhereyouarecollectinginformation
onaone-to-onebasis.Wheretheinformationiscollectedinacollectiveformatsuchasfocusgroups,community forums or panel discussions, you strive to gather as diverse and asmuch information aspossible.When no new information is emerging it is assumed that you have reached the saturationpoint.
SummaryInthischapteryouhavelearntaboutsampling,theprocessofselectingafewelementsfromasamplingpopulation.Sampling,inaway,isatrade-offbetweenaccuracyandresources.Throughsamplingyoumakeanestimateabouttheinformationofinterest.Youdonotfindthetruepopulationmean.Twoopposingphilosophiesunderpintheselectionofsamplingunitsinquantitativeandqualitativeresearch.Inquantitativestudies
asampleissupposedtobeselectedinsuchawaythatitrepresentsthestudypopulation,whichisachievedthroughrandomisation.However, the selection of a sample in qualitative research is guided by your judgement as towho is likely to provide youwithcompleteanddiverseinformation.Thisisanon-randomprocess.Sample size does not occupy a significant place in qualitative research and it is determined by the data saturation pointwhile
collectingdatainsteadofbeingfixedinadvance.Inquantitativeresearch,samplingisguidedbythreeprinciples,oneofwhichisthatthegreaterthesamplesize,themoreaccurate
theestimateofthetruepopulationmean,giventhateverythingelseremainsthesame.Theinferencesdrawnfromasamplecanbeaffectedbyboththesizeofthesampleandtheextentofvariationinthesamplingpopulation.Sampling designs can be classified as random/probability sampling designs, non-random/non-probability sampling designs and
‘mixed’samplingdesigns.Forasampletobecalledarandomsample,eachelementinthestudypopulationmusthaveanequalandindependent chanceof selection.Three randomdesignswerediscussed: simple random sampling, stratified randomsampling andcluster sampling. The procedures for selecting a sample using these designswere detailed step by step. The use of the fishbowltechnique,thetableofrandomnumbersandspecificallydesignedcomputerprogramsarethreecommonlyusedmethodsofselectingaprobabilitysample.Therearefivenon-probabilitysamplingdesigns:quota,accidental,judgemental,expertandsnowball.Eachisusedforadifferent
purposeandindifferentsituationsinbothquantitativeandqualitativestudies.Inquantitativestudiestheirapplicationisunderpinnedbythesamplesizewhereasthedatasaturationpointdeterminesthe‘samplesize’inqualitativestudies.Systematic sampling is classified under the ‘mixed’ category as it has the properties of both probability and non-probability
samplingdesigns.Thelastsectionofthechapterdescribeddeterminantsof,andproceduresfor,calculatingsamplesize.Althoughitmightbeslightly
moredifficultforthebeginner,thiswasincludedtomakeyouawareofthedeterminantsinvolvedasquestionsrelatingtothisareaaresocommonlyasked.Inqualitativeresearch,thequestionofsamplesizeislessimportant,asyouraimistoexplore,notquantify,theextentofvariationforwhichyouareguidedbyreachingsaturationpointintermsofnewfindings.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.
Considertheimplicationsofselectingasamplebaseduponyourchoiceasaresearcherandhowyoucouldmakesurethatyoudonotintroducebias.Intheabsenceofasamplingframeforemployeesofalargeorganisation,whichsamplingdesignwouldyouusetoselectasampleof219people?Explainwhyyouwouldchoosethisdesignandtheprocessyouwouldundertaketoensurethatthesampleisrepresentative.Fromyourownareaofinterest,identifyexamplesofwhereclustersamplingcouldbeapplied.Whatdeterminessamplesizeinqualitativeresearch?Whatisthedatasaturationpointinqualitativestudies?
STEPVWritingaResearchProposal
Thisoperationalstepincludesonechapter:
Chapter13:Writingaresearchproposal
CHAPTER13HowtoWriteaResearchProposal
Inthischapteryouwilllearnabout:
ThepurposeofaresearchproposalinquantitativeandqualitativeresearchHowtostructurearesearchproposalHowtowritearesearchproposal
Keywords: conceptual framework, data analysis, data processing, hypothesis,limitations, literature review, research design, research problem, sampling, studydesign,studyobjectives,theoreticalframework,time-frame.
Theresearchproposalinquantitativeandqualitativeresearch
All research endeavours, in both qualitative and quantitative research, in every academic andprofessionalfieldareprecededbyaresearchproposal.Itinformsyouracademicsupervisororpotentialresearchcontractprovideraboutyourconceptualisationofthetotalresearchprocessthatyouproposetoundertake so that they can examine its validity and appropriateness. In any academic field, yourresearchproposalwillgothroughanumberofcommitteesforapproval.Unlessitisapprovedbyallofthem,youwillnotbeabletostartyourresearch.Hence,it is importantforyoutostudycloselywhatconstitutesaresearchproposal.Youneedtowritearesearchproposalwhetheryourresearchstudyisquantitativeorqualitativeandin
both cases you use a similar structure. The main difference is in the proposed procedures andmethodologiesforundertakingtheresearchendeavour.Whenprovidingdetailsfordifferentpartsoftheresearchproposal,forquantitativestudies,youwilldetailquantitativemethods,proceduresandmodelsand,forqualitativestudies,yourproposedprocesswillbebaseduponmethodsandproceduresthatformthequalitativeresearchmethodology.Certain requirements for a research proposal may vary from university to university, and from
disciplinetodisciplinewithinauniversity.Whatisoutlinedherewillsatisfymostrequirementsbutyoushouldbeselectiveregardingwhatisneededinyoursituation.Aresearchproposalisanoverallplan,scheme,structureandstrategydesignedtoobtainanswersto
the research questions or problems that constitute your research project. A research proposal should
outlinethevarioustasksyouplantoundertaketofulfilyourresearchobjectives,testhypotheses(ifany)orobtainanswerstoyourresearchquestions.Itshouldalsostateyourreasonsforundertakingthestudy.Broadly,aresearchproposal’smainfunctionistodetailtheoperationalplanforobtaininganswerstoyour research questions. In doing so it ensures and reassures the reader of the validity of themethodologyforobtaininganswerstoyourresearchquestionsaccuratelyandobjectively.In order to achieve this function, a research proposalmust tell you, your research supervisor and
reviewersthefollowinginformationaboutyourstudy:
whatyouareproposingtodo;howyouplantofindanswerstowhatyouareproposing;whyyouselectedtheproposedstrategiesofinvestigation.
Contentsofaresearchproposal
Aresearchproposalshouldcontainthefollowinginformationaboutyourstudy:
anintroduction,includingabriefliteraturereview;theoreticalframeworkthatunderpinsyourstudy;conceptualframeworkwhichconstitutesthebasisofyourstudy;objectivesorresearchquestionsofyourstudy;hypothesestobetested,ifapplicable;studydesignthatyouareproposingtoadopt;settingforyourstudy;researchinstrument(s)youareplanningtouse;samplingdesignandsamplesize;ethicalissuesinvolvedandhowyouproposetodealwiththem;dataprocessingprocedures;proposedchaptersofthereport;problemsandlimitationsofthestudy;proposedtime-framefortheproject.
Aresearchproposalshouldcommunicate theabovecontentsclearlyandspecifically insuchawaythat anyonegoing through it shouldbeable toundertakeall tasks in the samemannerasyouwouldhave.Itshouldalso:
enableyoutoreturntotheproposalforyourownguidanceindecisionmakingatdifferentstagesoftheresearchprocess;convinceyourresearchsupervisororareviewerthatyourproposedmethodologyismeritorious,valid,appropriateandworkableintermsofobtaininganswerstoyourresearchquestionsorobjectives.
Universitiesandotherinstitutionsmayhavedifferingrequirementsregardingthestyleandcontentofaresearchproposal.Requirementsmayalsovarywithinaninstitution,fromdisciplinetodisciplineorfrom supervisor to supervisor. (The guidelines set out in this chapter therefore provide a frameworkwithinwhicharesearchproposalshouldbewritten.)
Yourproposal should follow the suggestedguidelinesandbewritten inanacademic style. Itmustcontainappropriatereferencesinthebodyofthetextandabibliographyattheend.Yoursurveyoftherelevant literatureshouldcovermajorpublicationson the topic.Thetheoretical framework foryourstudymustemergefromthisliteraturereviewandmusthaveitsgroundinginempiricalevidence.Asarule,theliteraturereviewincludes:
aconceptualframework,andtheoreticalandempiricalinformationaboutthemainissuesunderstudy;someofthemajorresearchfindingsrelatingtoyourtopic,researchquestionsraisedintheliteratureandgapsidentifiedbypreviousresearchers.
Your literature review should also raise issues relating to themethodologyyou are proposing.Forexample, itmay examine howother studies operationalised themajor variables of relevance to yourstudyandmayincludeacritiqueofmethodologyrelevanttoyourstudy.Thecritiquesofmethodsandproceduresshouldbeincludedundertheirrespectiveheadings.Forexample,acritiqueofthesamplingdesign you adopt should be included under ‘sampling’ or a critique to the study design should bediscussedunder‘studydesign’.Notethatthesuggestedresearchproposalstructuredoesnotcontainasectionentitled‘surveyofthe
literature’or ‘literature review’.This isbecause references to the literature shouldbe integratedwithyourargumentsconceptuallyratherthanchronologicallyandshouldbecomeapartofalltheaspectsofyourresearchreportfromproblemconceptualisationtoconclusions.Theliteratureshouldbereviewedunder main themes that emerge from your reading of the literature and should be included in the‘introduction’and‘theproblem’.Issuesidentifiedintheliteraturetodowithresearchmethodologyandproblems pertinent to the various aspects of research procedures should be discussed under theirrespective headings. For example, issues pertaining to the study design under ‘study design’, issuesrelatingtosamplingunder‘sampling’andtheliteraturepertainingtotheresearchinstrumentunderthe‘measurementprocedure’.In suggesting this format it is assumed that you are reasonably well acquainted with research
methodologyandanacademicstyleofwriting.That is,youknowhowtowriteasetofobjectivesorconstructahypothesis,youarefamiliarwiththevariousstudydesignsandyoucanconstructaresearchinstrumentandciteareference.Thepagesthatfollowoutlineaframeworkforaresearchproposal.Thecontentsundereachheading
mayvarymarkedly fromdiscipline todiscipline, according to theacademic levelof the student (BAHons,MA,PhD)andwhetheryourstudyispredominantlyquantitativeorqualitative.Forquantitativeproposalsyouneed tobeveryspecific inproposinghowyouaregoing toundertakeeachstepof theresearch journey, whereas for qualitative research proposals such details are not expected as yourmethodology is flexible and unstructured to accommodate in-depth search. However, you need toprovideabroadapproachtoyourenquiryasapartofyourresearchproposal.Eachsectionoftheproposedoutlineforaresearchproposalisdividedintotwoparts:
1. asuggestedtitleforthesectionandanoutlineofitscontents;2. examplesoutliningcontentsforthesection–thesamefourexamplesofresearchprojects,each
takenfromadifferentdiscipline,areusedasillustrationsineachsection.
Preamble/introduction
The proposal should start with an introduction to include some of the information listed below.Rememberthatsomeofthecontentssuggestedinthissectionmaynotberelevanttocertainstudies,souseyourdiscretioninselectingonlywhatispertinenttoyourstudy.Inwritingthissection,theliteraturereview (see Chapter 3 on reviewing the literature) is of central importance as it serves two mainfunctions:
1. Itacquaintsyouwiththeavailableliteratureintheareaofyourstudy,therebybroadeningyourknowledgebase.
2. Itprovidesyouwithinformationonthemethodsandproceduresotherpeoplehaveusedinsimilarsituationsandtellsyouwhatworksandwhatdoesnot.
Thetype,extentandqualityofaliteraturereviewaremostlydependentupontheacademiclevelforwhichyouarewritingtheproposal.Thecontentsofthissectionmayalsovarygreatlyaccordingtothesubjectareaunderstudy.Startwithaverybroadperspectiveofthemainsubjectarea,beforegraduallynarrowingthefocusto
thecentralproblemunderinvestigation.Indoingso,coverthefollowingaspectsofyourstudyarea:
anoverviewofthemainareaunderstudy;ahistoricalperspective(development,growth,etc.)pertinenttothestudyarea;philosophicalorideologicalissuesrelatingtothetopic;trendsintermsofprevalence,ifappropriate;majortheories,ifany;themainissues,problemsandadvancesinthesubjectareaunderstudy;importanttheoreticalandpracticalissuesrelatingtothecentralproblemunderstudy;themainfindingsrelatingtothecoreissue(s).
Fourexamplesofpossibletopicsforthepreamble/introductionforaresearchproposalfollow.
ExampleASupposethatyouareconductingastudytoinvestigatetheimpactofimmigrationonthefamily.Thepreamble/introductionshouldincludeabriefdescriptionofthefollowing:
Theoriginsofmigratorymovementsintheworld.Generaltheoriesdevelopedtoexplainmigratorybehaviour.Thereasonsformigration.Currenttrendsinmigration(nationalandstate).Theimpactofimmigrationonfamilyrolesandrelationships(e.g.onhusbandandwife,onchildrenandparents,onparentalexpectationsofchildren,etc.).Occupationalmobility.etc.
ExampleB
Suppose your research project is to conduct a study of the attitudes of foster carers towards foster payment in… (name of theplace/state/country).Thepreamble/introductionwouldincludethefollowing:
Theoriginsoffosterplacement,thephilosophyoffostercare,ahistoricaloverviewoffostercareandchangesovertheyears.Reasonsforfostercareandchangesovertime.Theoriginsoffosterplacementin…(thecountryinwhichyouareconductingyourstudy).Theeffectsoffosterplacementonchildrenandparents.Policieswithrespecttofostercarein…(theregion).Theoriginsoffostercarein…(theregion).Administrativeproceduresforfostercarein…(theregion).Thetrainingoffosterparentsin…(theregion).Theroleandresponsibilityoffosterparents.etc.
ExampleCSupposethatyouplantostudytherelationshipbetweenacademicachievementandsocialenvironment.Thepreamble/introductionwouldincludethefollowing:
Theroleofeducationinoursociety.Majorchangesinthephilosophyofeducationovertime.Factorsaffectingattitudestowardseducation.Thedevelopmentofeducationin…(country).Trendsineducationparticipationratesin…(country)withparticularreferencetotheregioninwhichthestudyisbeingcarriedout.Changingeducationalvalues.Roleofparentsandpeersinacademicachievement.Impactofsocialenvironmentonacademicachievement.etc.
ExampleDSuppose you are undertaking a qualitative study to find out what it means to have a child with ADHD in the family. Thepreamble/introductionshould includeyour thoughtsandarguments,andwhat the literaturesaysaround the followingaspectsofADHD.
DefinitionsandsymptomsofADHD.CausesofADHD.MedicalperspectiveonADHD.EffectsofADHDonfamilylife.TreatmentforADHD.Implicationsforachildifuntreated.
ManagementofADHD.etc.
Theproblem
Havingprovidedabroadintroductiontotheareaunderstudy,nowfocusonissuesrelatingtoitscentraltheme, identifying some of the gaps in the existing body of knowledge. Identify some of the mainunansweredquestions.Heresomeofthemainresearchquestionsthatyouwouldliketoanswerthroughyourstudyshouldalsoberaised,andarationaleandrelevanceforeachshouldbeprovided.Knowledgegainedfromotherstudiesandtheliteratureabouttheissuesyouareproposingtoinvestigateshouldbeanintegralpartofthissection.Specifically,thissectionshould:
identifytheissuesthatarethebasisofyourstudy;specifythevariousaspectsof/perspectivesontheseissues;identifythemaingapsintheexistingbodyofknowledge;raisesomeofthemainresearchquestionsthatyouwanttoanswerthroughyourstudy;identifywhatknowledgeisavailableconcerningyourquestions,specifyingthedifferencesofopinionintheliteratureregardingthesequestionsifdifferencesexist;developarationaleforyourstudywithparticularreferencetohowyourstudywillfilltheidentifiedgaps.
Thefollowingexamplesoutlinethetopicsaboutwhichtheliteratureshouldbereviewedandincludedinthesectionentitled‘Theproblem’.Keepinmindthatthesearejustsuggestionsandshouldserveonlyasexamplesforyoutodevelopandchangeasyoufeelappropriateforyourownstudy.
ExampleA
Whatsettlementprocessdoesafamilygothroughafterimmigration?Whatadjustmentsdoimmigrantshavetomake?Whattypesofchangecanoccurinfamilymembers’attitudes?(Theoryofacculturationetc.)Whatisthepossibleimpactofsettlementonfamilyrolesandrelationships?Intermsofimpact,whatspecificquestionsdoyouwanttoanswerthroughthestudy?Whatdoestheliteraturesayaboutthesequestions?Whatarethedifferentviewpointsontheseissues?Whatareyourownideasaboutthesequestions?Whatdoyouthinkwillbetherelevanceofthefindingsofyourstudytotheexistingbodyofknowledgeandtoyourprofession?Howwillthefindingsaddtothebodyofknowledgeandbeusefultoprofessionalsinyourfield?etc.
ExampleB
Whatarethebroadissues,debates,argumentsandcounter-argumentsregardingfoster-carepayment?Whataretheattitudesoffosterparentstotheamount,modeandformofpaymentandwhatdoestheliteraturesayabouttheseissues?Whatarethedifferentviewpoints/perspectivesregardingpaymentforfostercare?Whatmainquestionswillyourstudyanswer?Howwillyourfindingshelpinpolicyformulationandprogrammedevelopment?etc.
ExampleC
Whattheorieshavebeendevelopedtoexplaintherelationshipbetweenacademicachievementandsocialenvironment?Whatistherelationshipbetweeneducationalachievementandsocialenvironment:whattheoreticalmodelwillbethebasisofyourstudy?Whatdoprevioustheoriesandresearcheshavetosayregardingthecomponentsofthetheoreticalmodelandacademicachievement?Forexample,therelationshipbetweenacademicachievementand:
—theself-esteemandaspirations/motivationofastudent;—peergroupinfluence;—parentalinvolvementanditsrelationshipwiththeirsocioeconomicstatus;—themotivationandinterestofstudentsinthesubject;—employmentprospects;—relationshipwithateacher;—etc.
ExampleD
WhataretheeffectsonthefamilyofhavingachildwithADHDinthefamilyasidentifiedintheliterature?Accordingtotheliterature,arethereanydifferencesbetweentheseeffectsandthetypeoffamily?WhatstrategieshavebeenusedforthemanagementofADHDbyafamily?Whateffects,accordingtotheliterature,doesADHDhaveonsiblingrelationships?WhataretheperceptionsoffamilymembersabouttheeffectsandmanagementofADHD?HowdofamiliescopewhentheyhaveachildwithADHDinthefamily?etc.
Objectivesofthestudy
In thissectionincludeastatementofbothyourstudy’smainandsubobjectives(seeChapter4).Yourmainobjectiveindicatesthecentralthrustofyourstudywhereasthesubobjectivesidentifythespecificissuesyouproposetoexamine.Theobjectivesofthestudyshouldbeclearlystatedandspecificinnature.Eachsubobjectiveshould
delineate only one issue. Use action-oriented verbs such as ‘to determine’, ‘to find out’ and ‘toascertain’informulatingsubobjectives,whichshouldbenumericallylisted.Iftheobjectiveistotestahypothesis, you must follow the convention of hypothesis formulation in wording the specificobjectives.In qualitative studies the statement of objectives is not as precise as in quantitative studies. In
qualitativestudiesyoushouldsimplymentionanoverallobjectiveofthestudyasyouraimistoexploreasmuchaspossibleasyougoalong.Asyouknow,thestrengthofqualitativeresearchisinflexibilityofapproachand theability to incorporatenew ideaswhilecollectingdata.Havingstructuredstatementsthatbindyoutoapredeterminedframeworkofexplorationisnotapreferredconventioninqualitativeresearch.Statements like to explore ‘whatdoes itmean tohave a childwithADHD in the family?’,‘how does it feel to be a victim of domestic violence?’, ‘how do people cope with racialdiscrimination?’, ‘the relationshipbetween resilienceandyoga’or ‘reconstructing lifeafterbushfire’,aresufficienttocommunicateyourintentofobjectivesinqualitativeresearch.Moredetailedobjectives,ifneedbe,canbedevelopedafterastudyiscomplete.
ExampleAMainobjective:Toascertaintheimpactofimmigrationonthefamily.
Subobjectives:
1. Todeterminetheimpactofimmigrationonhusband/wiferolesasperceivedbyimmigrants.2. Tofindouttheimpactofimmigrationonmaritalrelations.3. Toascertainperceivedchangesinparentalexpectationsofchildren’sacademicand
professionalachievement.4. Todetermineperceivedchangesofattitudetowardsmarriageinthestudypopulation.
ExampleBMainobjective:Todeterminetheopinionoffostercarersabouttheformandextentoffosterpaymenttheyfeeltheyshouldreceivefortakingcareofafosterchild.
Subobjectives:
1. Todeterminetheformandmodeofpaymentfortakingcareofafosterchild.2. Toidentifythefactorsthatfosterparentsbelieveshouldbethebasisfordeterminingtherate
ofpaymentforfosteringachild.3. Todeterminetherelationship,ifany,betweenthesocioeconomicgraphiccharacteristicsof
fosterparentsandtheirviewsonpayment.
ExampleCMainobjective:Toexaminetherelationshipbetweenacademicachievementandsocialenvironment.
Subobjectives:
1. Tofindouttherelationship,ifany,betweenself-esteemandastudent’sacademicachievementatschool.
2. Toascertaintheassociationbetweenparentalinvolvementinastudent’sstudiesandhis/heracademicachievementatschool.
3. Toexaminethelinksbetweenastudent’speergroupandacademicachievement.4. Toexploretherelationshipbetweenacademicachievementandtheattitudeofastudent
towardsteachers.
ExampleDMainobjective:ToexplorewhatitmeanstohaveachildwithADHDinthefamily.
Hypothesestobetested
A hypothesis is a statement of your assumptions about the prevalence of a phenomenon or about arelationshipbetweentwovariablesthatyouplantotestwithintheframeworkofthestudy(seeChapter6).Ifyouaregoingtotesthypotheses,listtheminthissection.Whenformulatingahypothesisyouhaveanobligationtodrawconclusionsaboutitinthetextofthe
report.Hypotheseshaveaparticularstyleofformulation.Youmustbeacquaintedwiththecorrectwayofwordingthem.Inastudyyoumayhaveasmanyhypothesesasyouwanttotest.However,itisnotessentialtohaveahypothesisinordertoundertakeastudy–youcanconductaperfectlysatisfactorystudywithoutformulatingahypothesis.
ExampleAH1=Inmostcasestherewillbeachangeinhusband/wiferolesafterimmigration.
H2=Inamajorityofcasestherewillbeachangeinparents’expectationsoftheirchildren.
Hi=etc.
ExampleBH1=Mostpeoplebecomefosterparentsbecauseoftheirloveofchildren.
H2=Amajorityoffosterparentswouldliketobetrainedtocareforfosterchildren.Hi=etc.
ExampleCH1=Astudent’sself-esteemandacademicachievementatschoolarepositivelycorrelated.
H2 = The greater the parental involvement in a student’s studies, the higher the academicachievement.H3 = A student’s attitude towards teachers is positively correlated with his/her academicachievementinthatsubject.Hi=etc.
ExampleDHypothesesarenotconstructedinqualitativeresearch.
Studydesign
Describethestudydesign(fordetailsseeChapter8)youplantousetoansweryourresearchquestions.(Forexample,saywhetheritisacasestudy,descriptive,cross-sectional,before-and-after,experimentalornon-experimentaldesign.)Identifythestrengthsandweaknessesofyourstudydesign.Includedetailsaboutthevariouslogisticalproceduresyouintendtofollowwhileexecutingthestudy
design.Onecharacteristicofagoodstudydesignisthatitexplainsthedetailswithsuchclaritythat,ifsomeoneelsewantstofollowtheproposedprocedure,s/hewillbeabletodoexactlyasyouwouldhavedone.Yourstudydesignshouldincludeinformationaboutthefollowing:
Whomakesupthestudypopulation?Caneachelementofthestudypopulationbeidentified?Ifyes,how?Willasampleorthetotalpopulationbestudied?Howwillyougetintouchwiththeselectedsample?Howwillthesample’sconsenttoparticipateinthestudybesought?Howwillthedatabecollected(e.g.byinterview,questionnaireorobservation)?Inthecaseofamailedquestionnaire,towhataddressshouldthequestionnairebereturned?Areyouplanningtosendareminderregardingthereturnofquestionnaires?Howwillconfidentialitybepreserved?
Howandwherecanrespondentscontactyouiftheyhavequeries?
ExampleAThestudy isprimarilydesigned to findout fromacross-sectionof immigrants from…,…and…(namesof thecountries) theperceived impact of immigration on family roles. Initial contact with the ethnic associations for these countries will be madethroughtheelectedofficebearerstoobtainalistofmembers.Fiveimmigrantswillbeselectedfromthelistatrandom,andwillbecontactedbyphone toexplain thepurposeof thestudyand its relevance,and toseek theiragreement toparticipate in thestudy.Those who give their consent will be interviewed at their homes or any other convenient place. To select a further sample, asnowballsamplingtechniquewillbeuseduntilthedesiredsamplesizeisobtained.
ExampleBThestudydesigniscross-sectionalinnature,beingdesignedtofindoutfromacross-sectionoffosterparentstheiropinionsaboutfoster payment. All foster parents currently registered with the Department of … (name of the office) constitute the studypopulation.Fromtheexistingrecordsofthisdepartmentitseemsthatthereare457fosterparentsin…(nameoftheregion).Asitisimpossible for the researcher,within the constraints of time andmoney, to collect information from all the foster parents, it isproposedtoselectasampleof50percentofthestudypopulationwiththeproposedsamplingstrategy.Thequestionnaire,withasupporting letter from the department will be sent with a prepaid envelope. The respondents will be requested to return thequestionnaireby…(date).Theletterfromtheresearcherattachedtothequestionnairewillexplaintheobjectivesandrelevanceofthestudy,assure therespondentsofanonymityandgive themtheoptionofnotparticipatingin thestudyif theywish.Acontactnumberwill be provided in case a respondent has any questions. In the case of a low response rate (less than 25 per cent), areminderwillbesenttorespondents.
ExampleCIt is proposed that the studywill be carried out in twogovernment high schools in themetropolitan area.Theprincipals of theschoolsmostaccessibletotheresearcherwillbecontactedtoexplainthepurposeofthestudyandthehelpneededfromtheschool,andtoseektheirpermissionforthestudentstoparticipateinthestudy.Astheconstraintsoftimeandresourcesdonotpermittheresearchertoselectmorethantwoschools,negotiationswithotherschoolswillceasewhentwoschoolsagreetoparticipateinthestudy.ItisproposedtoselectYear9studentsastheacademicachievementofstudentsinYears8and10couldbeaffectedbyfactors
uniquetothem.Year8studentsmaybeexperiencinganxietyasaresultofhavingjustmadethetransitiontoanewsystem.ThemotivationofstudentsinYear10couldbeaffectedbytheirbeingatthestageintheireducationwheretheymustdecideiftheywillstayonatschool.Inordertocontrolthevarianceattributabletothegenderofastudentitisproposedtoselectonlymalestudents.Oncetheprincipalofaschoolagreestoallowthestudytobecarriedout,theresearcherwillbrieftheteacherinchargeaboutthe
studyanditsrelevance,andwillarrangeadateandtimeforadministeringthequestionnaire.Whenthestudentsareassembled,readytoparticipateinthestudy,theresearcherwillexplainitspurposeandrelevance,andthen
distributethequestionnaire.Theresearcherwillremainwiththeclasstoansweranyquestionsthestudentsmighthave.
ExampleDTheresearcherisknowntoafamilythathasachildwithADHDandthatbelongstoanADHDsupportgroupwhichmeetseverymonth.Theresearcherproposestomakeinitialcontactwiththegroupthroughtheknownfamily.Theresearcherwillattendoneofthemonthlymeetingsandbriefthegrouponthepurposeandrelevanceofthestudy,criteriaforinclusioninthestudy,whatitentailstobeinvolvedinthestudy,andotheraspectsofthestudy.Therespondentswillalsobeassuredoftheanonymityoftheinformationsharedbythemanditsethicaluse.Themembersofthegroupwillbeencouragedtoaskquestionsaboutanyaspectofthestudy.Havingsoughttheirconsent,theresearcherwillseekopinionsofsomegroupmemberstodecidewhoshouldparticipateinthestudyinlightoftheinclusioncriteria.Itisproposedtoselectsixfamilies,threewherebothparentsareinvolvedinthetreatmentandmanagementofanADHDchild
andthreefromfamilieswherethemotheristhesolecarer.Thisisprimarilytoseeiftherearedifferencesinlookingafterachild
withADHDamongdifferenttypesoffamily.Thepotential respondentswill be individually contactedby the researcher to seek their consent for participation in the study.
Onceconsenthasbeenobtainedtheplaceandtimingsforinterviewswillbefixedwitheachfamily.Dependinguponthetypeoffamily, the issues will be discussed either collectively with the father and mother or with the mother only. Before starting aninterviewtheirpermissiontorecordtheinterviewonataperecorderwillbesought.Havingcompletedtheinterviews,theresearcherwilltranscribetheresponsesandacopywillbegiventotherespondentsforconfirmationandvalidation.
Thesetting
Brieflydescribe theorganisation, agencyor community inwhichyouwill conductyour study. If thestudy is about a group of people, highlight some of the salient characteristics of the group (e.g. itshistory,size,compositionandstructure)anddrawattentiontoanyavailablerelevantinformation.Ifyourresearchconcernsanagency,officeororganisation,includethefollowinginyourdescription:
themainservicesprovidedbytheagency,officeororganisation;itsadministrativestructure;thetypeofclientsserved;informationabouttheissuesthatarecentraltoyourresearch.
Ifyouarestudyingacommunity,brieflydescribesomeofthemaincharacteristics,suchas:
thesizeofthecommunity;abriefsocialprofileofthecommunity(i.e.thecompositionofthevariousgroupswithinit);issuesofrelevancetothecentralthemeofyourstudy.
Notethat,duetothenatureofthecontent,itwouldbedifficulttoprovideexamples.
Measurementprocedures
Thissectionshouldcontainadiscussionofyourinstrument(seeChapters9and10)andthedetailsofhowyouplantooperationaliseyourmajorvariables(Chapter5).To start with, justify your choice of research tool, highlighting its strengths and pointing out its
weaknesses. Then outline themajor segments of your research tool and their relevance to themainobjectives of the study. If you are using a standard instrument, briefly discuss the availability ofevidenceonitsreliabilityandvalidity.Ifyouadaptormodifyit inanyway,describeandexplainthechangesyouhavemade.You should also discuss how you are going to operationalise themajor concepts. For example, if
measuringeffectiveness,specifyhowitwillbemeasured.Ifyouplantomeasuretheself-esteemofagroup of people,mention themain indicators of self-esteem and the procedures for itsmeasurement(e.g.theLikertorThurstonescale,oranyotherprocedure).Ideally,forquantitativestudiesyoushouldattachacopyoftheresearchinstrumenttoyourproposal.Notethat,duetothenatureofthecontent,itwouldbedifficulttoprovideexamplesforthissection.
Ethicalissues
Allacademic institutionsareparticularaboutanyethical issues that researchmayhave.Todealwiththem, all institutions have some form of policy on ethics. You need to be acquainted with yourinstitution’spolicy. It is imperative that inyourproposalyou identifyanyethical issuesanddescribehow you propose to deal with them. You need to look at the ethical issues particularly from theviewpoint of your respondents and, in case of any potential ‘harm’, psychological or otherwise, youneedtodetailthemechanisminplacetodealwithit.FurtherinformationonethicalissuesisprovidedinChapter14.
Sampling
Underthissectionoftheproposalincludethefollowing(consultChapter12onsampling):
thesizeofthesamplingpopulation(ifknown)andfromwhereandhowthisinformationwillbeobtained;thesizeofthesampleyouareplanningtoselectandyourreasonsforchoosingthissize;anexplanationofthesamplingdesignyouareplanningtouseintheselectionofthesample(simplerandomsampling,stratifiedrandomsampling,quotasampling,etc.).
ExampleABecausealackofinformationastotheexactlocationofmigrantfamiliesmakesitdifficulttouseaprobabilitysamplingdesign,itisproposedthattheresearcherwillemployasnowballsamplingtechnique.Theresearcherwillmakeinitialcontactwithfivefamilieswhohaveemigratedfrom…(nameofthecountry)duringthepastseventotenyears,whoareeitherknowntohim/heroronthebasis of information obtained from the office bearers of the formal associations representing the migrant groups. From eachrespondenttheresearcherwillobtainnamesandaddressesofotherimmigrantswhohavecomefromthesamecountryduringthesameperiod.Therespondents thus identifiedwill thenbe interviewedandasked to identifyother respondents for theresearcher.Thisprocesswillcontinueuntiltheresearcherhasinterviewed70respondents.
ExampleBBecauseoftheconstraintsoftimeandresourcesitisproposedtoselect50percentofthefosterparentscurrentlyregistered(457)withthedepartmentusingthesystematicrandomsamplingtechnique.Everyotherfosterparentregisteredwiththedepartmentwillbeselected,thus229individualswillconstitutethesampleforthestudy.
ExampleCTheselectionofschoolswillbedoneprimarilythroughquotasampling.Schoolswillbeselectedonthebasisoftheirgeographicalproximitytotheresearcher.Theresearcherwillpreparealistofschools,inrankorder,ofaccessibility.Oncetwoschoolsagreetoparticipateinthestudy,negotiationswithotherschoolswillcease.AllYear9malestudentswillconstitutethestudypopulation.Itisexpectedthatthesamplewillnotexceed100students.
ExampleDItisproposedtousethejudgemental/purposivesamplingtechniquetoselectsixfamiliesfromthegroup,threewherebothparentslookafteranADHDchildand threewhereonly themotherhas themain responsibility (singleparent families).On thebasisof
informaldiscussionswiththegroupmembers,thosefamilieswhoareexpectedtobeinformationrichintreatingandmanagingachildwithADHDwillbeselectedtobeinterviewed.
Analysisofdata
Ingeneralterms,describethestrategyyouintendtousefordataanalysis(Chapter15).Specifywhetherthedatawill be analysedmanuallyor by computer.For computer analysis, identify theprogramandwhereappropriate the statisticalproceduresyouplan toperformon thedata.Forquantitative studiesalsoidentifythemainvariablesforcross-tabulation.For qualitative studies, describe how you plan to analyse your interviews or observation notes to
drawmeaningsfromwhatyourrespondentshavesaidaboutissuesdiscussedorobservationnotesmade.One of the common techniques is to identify main themes, through analysing the contents of theinformation gathered by you in the field.You first need to decidewhether youwant to analyse thisinformationmanuallyoruseacomputerprogramforthepurpose.Therearethreewaystoproceedwithcontentanalysis:
1. Fromyourfieldnotesdevelopaframeworkofyourwrite-upandasyougothroughyournotesdirectlyintegratethatinformationwithinthestructuredeveloped.Ifyouadoptthismethod,youneedtobereasonablyclearaboutthestructure.Itdoesnotmeanthatyoucannotdevelopthestructureasyougoonanalysing;still,aclearvisionwillbeofimmensehelpinslottinginformationgatheredinthefieldbyyouintothewrite-up.
2. Thesecondmethodisthatyoutranscribeyourfieldnotestobereadbyyouoverandoveragaintoidentifythemainthemes.Thesethemesbecomethebasisofyourwrite-up.
3. TherearecomputerprogramssuchasNUD*IST,Ethnograph,NVivospecificallydesignedtohandledescriptivedata.Youmayprefertouseoneoftheseprograms.Theseprogramsarealsobasedupontheprincipleofcontentanalysis.Theonlydifferenceisthatinsteadofyoursearchingmanually,theyidentifywhereaparticulartextidentifyingthethemeappears.
Youneedtospecifywhichparticularstrategyyouareproposingfordataanalysisforyourstudy.
ExampleAFrequencydistributionsintermsof:
age;education;occupation;numberofchildren;durationofimmigration;etc.
Cross-tabulations:Impactofhusband/wiferoles
age;
numberofchildren;education;occupation;etc.
ExampleBFrequencydistributionsintermsof:
age;income;education;occupation;maritalstatus;durationoffostercare;numberoffosterchildren;etc.
Cross-tabulations:Attitudetowardsfosterpayment
age;numberofchildren;education;occupation;etc.
Statisticalteststobeapplied:
chisquare;regressionanalysis;etc.
ExampleCFrequencydistributionsintermsof:
age;parents’occupation;parents’educationallevels;students’occupationalaspirations;
parentalinvolvementinstudents’studies;self-esteem;peergroupinfluence;numberofhoursspentonstudies;etc.
Cross-tabulations:Academicachievement
peergroupinfluence;parentalinvolvementinstudents’studies;self-esteem;occupationalaspirations;attitudetowardsteachers;etc.
ExampleDThein-depthinterviewscarriedoutwiththefamilieswillbetranscribedusingMicrosoftWord.Thesetranscribedinterviewswillbecloselystudiedtoidentifythemainthemestheycommunicate.ThesethemeswillbesortedbyissuesrelatingtomanagementandtreatmentofachildwithADHD.Thethemeswillthenbecomepartofthewrite-up.
Structureofthereport
Asclearlyaspossible,statehowyouintendtoorganisethefinalreport(seeChapter17).Inorganisingyourmaterialforthereport,thespecificobjectivesofyourstudyareofimmensehelp.Plantodevelopyour chapters around the main themes of your study. The title of each chapter should clearlycommunicatethemainthrustofitscontents.The first chapter, possibly entitled ‘Introduction’, should be an overall introduction to your study,
coveringmostofyourprojectproposalandpointingoutdeviations,ifany,fromtheoriginalplan.Thesecondchaptershouldprovidesomeinformationaboutthestudypopulationitself–thatis,some
ofitssocioeconomic–demographiccharacteristics.Themainaimofthischapteristogivereaderssomebackgroundonthepopulationfromwhichyoucollectedtheinformation.Thesecondchapter,therefore,may be entitled, ‘Socioeconomic–demographic characteristics of the study population’ or ‘The studypopulation’oranyothertitlethatcommunicatesthisthemetoreaders.Titlesfortherestofthechapterswillvaryfromstudytostudybut,asmentioned,eachchaptershouldbewrittenaroundamaintheme.Althoughthewordingofchaptertitlesisanindividualchoice,eachmustcommunicatethemainthemeofthechapter.Indevelopingthesethemesthespecificobjectivesofthestudyshouldbekeptinthefrontofyourmind.If your study is qualitative, the main issues identified during data collection and analysis stages
shouldbecomethebasisofdevelopingchapters.Havingdevelopedsignificantissues,thenextstepistoorganisethemainthemesundereachissueanddevelopastructurethatyouwillfollowtocommunicateyourfindingstoyourreaders.
ExampleAItisproposedthatthereportwillbedividedintothefollowingchapters:Chapter1: IntroductionChapter2: Thesocioeconomic–demographiccharacteristicsofthestudypopulationChapter3: Theimpactonhusband/wiferolesChapter4: TheimpactonmaritalrelationsChapter5: TheimpactonexpectationsofchildrenChapter6: TheimpactonattitudestowardsmarriageChapter7: Summary,conclusionsandrecommendations
ExampleBThedissertationwillbedividedintothefollowingchapters:Chapter1: IntroductionChapter2: AprofileofthestudypopulationChapter3: Fostercarers’perceptionsoftheirroleChapter4: Attitudesoffostercarerstowardsfoster-carepaymentChapter5: ThepreferredmethodofpaymentChapter6: GeneralcommentsmadebyrespondentsaboutfostercareChapter7: Summary,conclusionsandrecommendations
ExampleCThereportwillhavethefollowingchapters:Chapter1: IntroductionChapter2: ThestudypopulationChapter3: Occupationalaspirations,self-esteemandacademicachievementChapter4: TheextentofparentalinvolvementandacademicachievementChapter5: PeergroupinfluenceandacademicachievementChapter6: AcademicachievementandstudentattitudestowardsteachersChapter7: Summary,conclusionsandrecommendations
ExampleDItisproposedthatthereportwillhavethefollowingchapters:Chapter1: ADHD:Atheoreticalperspective
Chapter2: IssuesanddifficultiesfacedbyfamilymembersinbringingupachildwithADHD
Chapter3: ADHDanditsperceivedeffectsonthechildChapter4: ADHDanditsperceivedimpactonsiblingrelationshipsChapter5: ManagingtreatmentChapter6: PerceivedeffectsofADHDonschoolingofthechildChapter7: PerceivedeffectsofADHDonrelationshipswithotherchildrenChapter8: AcasehistoryChapter9: Summaryandconclusions
Problemsandlimitations
This section should list any problems you think you might encounter concerning, for example, theavailabilityofdata,securingpermissionfromtheagency/organisationtocarryoutthestudy,obtainingthesample,oranyotheraspectofthestudy.Youwillnothaveunlimitedresourcesandasthismaybeprimarilyanacademicexercise,youmight
have todo less thanan ideal job.However, it is important tobeawareof– andcommunicate– anylimitationsthatcouldaffectthevalidityofyourconclusionsandgeneralisations.Here, problems refer to difficulties relating to logistical details, whereas limitations designate
structural problems relating tomethodological aspectsof the study. Inyouropinion the studydesignyou chosemay not be the best but youmight have had to adopt it for a number of reasons.This isclassifiedasalimitationofthestudy.Thisisalsotrueforsamplingormeasurementprocedures.Suchlimitationsshouldbecommunicatedtoreaders.
Appendix
Asanappendix,inthecaseofquantitativestudies,attachyourresearchinstrument.Also,attachalistofreferencesintheappendixoftheproposal.
Workschedule
Youmustsetyourselfdatesasyouneedtocompletetheresearchwithinacertaintime-frame.Listthevariousoperationalstepsyouneedtoundertakeandindicateagainsteachthedatebywhichyouaimtocomplete that task.Remember tokeepsome time towards theendasa ‘cushion’ incase theresearchprocessdoesnotgoassmoothlyasplanned.DevelopachartasshowninTable13.1.
TABLE13.1Developingatime-frameforyourstudy
SummaryAresearchproposaldetailstheoperationalplanforobtaininganswerstoresearchquestions.Itmusttellyoursupervisorandotherswhatyouproposetodo,howyouplantoproceedandwhythechosenstrategyhasbeenselected.Itthusassuresreadersofthevalidityofthemethodologyusedtoobtainanswersaccuratelyandobjectively.The guidelines set out in this chapter provide only a framework within which a research proposal for both quantitative and
qualitativestudiesshouldbewrittenandassumethatyouarereasonablywellacquaintedwithresearchmethodologyandanacademicstyle of writing. The contents of your proposal are arranged under the following headings: preamble/introduction, the problem,objectivesofthestudy,hypothesestobetested,studydesign,setting,measurementprocedures,sampling,analysisofdata,structureofthereport,andproblemsandlimitations.Thespecifics,undereachheading,willvarywiththetypeofstudyyouareproposingtoundertake. The write-up for qualitative studies will be based upon qualitative methodology and quantitative methodology willdeterminethecontentsofquantitativestudies.The ‘preamble’ or ‘introduction’ introduces themain area of the study.To startwith, the literature review is broad and then it
graduallynarrowstothespecificproblemyouareinvestigating.Thetheoreticalframeworkshouldbeapartofthissection.Thenextsection,‘theproblem’,detailsthespecificproblemunderstudy.Theresearchquestionsforwhichyouareplanningtofindanswersareraisedinthissection.‘Objectivesofthestudy’containsyourmainobjectivesandyoursubobjectives.Hypotheses,ifany,shouldbe listed in thesection‘hypotheses tobe tested’.The logisticalproceduresyou intend to followaredetailedunder ‘studydesign’.‘Thesetting’consistsofadescriptionoftheorganisationorcommunityinwhichyouplantoconductyourstudy.Theprocedureforobtaininginformationandthemeasurementofmajorvariablesareexplainedinthe‘measurementprocedures’section.Youneedtowriteaboutethical issuesthatyourstudymighthaveandhowyouproposetodealwiththem.Howyouwillselectyoursampleisdescribedunder‘sampling’.Theprocedurefordataanalysisisdiscussedunder‘analysisofdata’.Thewayyouplantostructureyourreport is outlinedunder ‘structureof the report’.Anticipatedproblems in conducting the studyand limitationswith its designaredescribedunder‘problemsandlimitations’.Asanappendixtoyourproposalattachacopyoftheresearchinstrumentandalistofthereferences.Aworkscheduleprovidesatime-frameforyourstudy.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Comparetheresearchproposalcontentssuggestedinthischapterwiththoserecommendedbyyouruniversityordepartment.Iftheyaredifferent,whatarethedifferences?Findouttheprocessthataresearchproposalgoesthroughinyouruniversitybeforeapprovalisgranted.
STEPVICollectingData
Thisoperationalstepincludesonechaptertomakeyouawareoftheethicalissuesinresearch:
Chapter14:Consideringethicalissuesindatacollection
CHAPTER14ConsideringEthicalIssuesinDataCollection
Inthischapteryouwilllearnabout:
Ethics:theconceptStakeholdersinresearchEthicalissuestoconsiderconcerningresearchparticipantsEthicalissuestoconsiderrelatingtotheresearcherEthicalissuestoconsiderregardingthesponsoringorganisation
Keywords: bias, codeof conduct, confidentiality, deprivationof treatment, ethos,harm, informed consent, principles of conduct, research participants, sensitiveinformation,sponsoringorganisations,stakeholders,subjectivity.
Ethics:theconcept
All professions are guided by a code of ethics that has evolved over the years to accommodate thechanging ethos, values, needs and expectations of those who hold a stake in the professions. Someprofessionsaremoreadvancedthanothersintermsofthelevelofdevelopmentoftheircodeofethics.Somehaveverystrictguidelines,monitorconducteffectivelyandtakeappropriatestepsagainstthosewhodonotabidebytheguidelines.Mostprofessionshaveanoverallcodeofconductthatalsogovernsthewaytheycarryoutresearch.
In addition, many research bodies have evolved a code of ethics separately for research.Medicine,epidemiology, business, law, education, psychology and other social sciences have well-establishedcodesofethicsforresearch.Let us first examine what we mean by ‘ethics’ or ‘ethical behaviour’. According to the Collins
Dictionary (1979: 502), ethicalmeans ‘in accordancewith principles of conduct that are consideredcorrect,especiallythoseofagivenprofessionorgroup’.Thekeywordshere,‘principlesofconduct’and‘consideredcorrect’,raisecertainquestions:
Whataretheseprinciplesofconduct?Whodeterminesthem?
Inwhosejudgementmusttheybeconsideredcorrect?
Closelyrelatedquestionsareasfollows:
Arethereuniversalprinciplesofconductthatcanbeappliedtoallprofessions?Dothesechangewithtime?Shouldthey?Whathappenswhenaprofessionaldoesnotabidebythem?
Thesubjectofethicsneedstobeconsideredinlightofthesequestions.Thewayeachprofessionservessocietyiscontinuouslychanginginaccordancewithsociety’sneeds
and expectations and with the technology available for the delivery of a service. The ethical codesgoverning themanner inwhichaservice isdeliveredalsoneed tochange.Whathasbeenconsideredethicalinthepastmaynotbesojudgedatpresent,andwhatisethicalnowmaynotremainsointhefuture.Anyjudgementaboutwhetheraparticularpracticeisethicalismadeonthebasisofthecodeofconductprevalentatthatpointintime.Astheserviceanditsmannerofdeliverydifferfromprofessiontoprofession,nocodeofconductcan
beuniformlyappliedacrossallprofessions.Eachprofessionhasitsowncodeofethics,thoughtherearecommonalities. If you want guidelines on ethical conduct for a particular profession, you need toconsultthecodeofethicsadoptedbythatprofessionordiscipline.‘Whataretheseprinciplesofconduct?’isthemostimportantquestionasitaddressestheissueofthe
contents of ethical practice in a profession. As the code of conduct varies from profession toprofession,itisnotpossibletoprovideauniversalanswertothisquestion.However,inresearch,anydilemmastemmingfromamoralquandaryisabasisofethicalconduct.Therearecertainbehavioursinresearch–suchascausingharmtoindividuals,breachingconfidentiality,usinginformationimproperlyandintroducingbias–thatareconsideredunethicalinanyprofession.The next question is: in whose judgement must a code of conduct be considered correct? Who
decideswhether a particular practice iswrong? If a procedure is carriedoutwrongly,what penaltiesshouldbeimposed?Itistheoverallbodyofprofessionalsorgovernmentorganisationsthatcollectivelydevelops a professional code of conduct and forms a judgement as to whether or not it is beingfollowed.Asmentioned,mostprofessionshaveestablishedanoverallcodeofethicsandalsoacodeofethics
forconductingresearchintheirrespectivefields.Asthisbookisdesignedforresearchersinthesocialsciences,wewillexamineethicalissuesrelatingtoresearchingeneralandissuesthatareapplicabletomostsocialsciencedisciplines.
Stakeholdersinresearch
Therearemany stakeholders inresearch,whether it is quantitativeorqualitative. It is important tolookatethicalissuesinrelationtoeachofthem.Thevariousstakeholdersinaresearchactivityare:
1. theresearchparticipantsorsubjects;2. theresearcher;3. thefundingbody.
Whoshouldbeconsideredasaresearchparticipantvariesfromprofessiontoprofession.Generally,allthosewithdirectandindirectinvolvementinaresearchstudyareconsideredasresearchparticipants,hencestakeholders.Inaddition,thosewhoarelikelytobeaffectedbythefindingsofastudyarealsoconsideredasstakeholders.Inthefieldsofmedicine,publichealth,epidemiologyandnursing,patientsandnon-patientswhobecomepart of a study and thosewhoparticipate in an experiment to test theeffectivenessofadrugor treatmentareconsideredasresearchparticipants.Serviceproviders,servicemanagers and planners who are involved in either providing the service or collecting informationrelatingtothestudyarealsostakeholdersintheresearch.Inthesocialsciences,theparticipantsincludeindividuals,groupsandcommunitiesprovidinginformationtohelparesearchertogainunderstandingof a phenomenon, situation, issue or interaction. In socialwork and psychology, participants includeclients as well as non-clients of an agency from whom information is collected to find out themagnitudeofaproblem,theneedsofacommunityortheeffectivenessofanintervention;andserviceproviders,socialworkersandpsychologists,whentheyprovideinformationforastudy.Inmarketing,consumersaswellasnon-consumersofaproductprovideinformationaboutconsumptionpatternsandbehaviour. In education, subjects include students, teachers andperhaps the community at largewhoparticipateineducationalresearchactivities.Similarly,inanydisciplineinwhicharesearchactivityisundertaken,thosefromwhominformationiscollectedorthosewhoarestudiedbyaresearcherbecomeparticipantsofthestudy.Researchersconstitutethesecondcategoryofstakeholders.Anyonewhocollectsinformationforthe
specific purpose of understanding, consolidation, enhancement and development of professionalknowledge,adheringtotheacceptedcodeofconduct,isaresearcher.S/hemayrepresentanyacademicdiscipline.Funding organisations responsible for financing a research activity fall into the third category of
stakeholders. Most research is carried out using funds provided by business organisations,pharmaceutical companies, service institutions (government, semi-governmentorvoluntary), researchbodiesand/oracademicinstitutions.Thefundsaregivenforspecificpurposes.Each category of stakeholders in a research activity may have different interests, perspectives,
purposes,aimsandmotivations thatcouldaffect thewayinwhich theresearchactivity iscarriedoutandthewayresultsarecommunicatedandused.Becauseofthis,itisimportanttoensurethatresearchisnotaffectedbytheself-interestofanypartyandisnotcarriedoutinawaythatharmsanyparty.Itistherefore important to examine ethical conduct in research concerning different stakeholders underseparatecategories.
Ethicalissuestoconsiderconcerningresearchparticipants
Therearemanyethicalissuestoconsiderinrelationtotheparticipantsofaresearchactivity.
Collectinginformation
One could ask: why should a respondent give any information to a researcher? What right does aresearcherhave toknockatsomeone’sdooror tosendoutaquestionnaire? Is itethical todisturbanindividual,evenifyouaskpermissionbeforeaskingquestions?Whyshouldapersongiveyouhis/hertime?Yourrequestforinformationmaycreateanxietyorputpressureonarespondent.Isthisethical?But the abovequestionsdisplay anaive attitude.The authorbelieves that if this attitudehadbeen
adopted, therewould have been no progress in theworld. Research is required in order to improve
conditions.Providedanypieceofresearchislikelytohelpsocietydirectlyorindirectly,itisacceptableto ask questions, if you first obtain the respondents’ informedconsent.Before you begin collectinginformation,youmustconsidertherelevanceandusefulnessoftheresearchyouareundertakingandbeable to convince others of this also. If you cannot justify the relevance of the research you areconducting,youarewastingyourrespondents’time,whichisunethical.
Seekingconsent
In every discipline it is considered unethical to collect information without the knowledge ofparticipants, and their expressed willingness and informed consent. Seeking informed consent ‘isprobably the most common method in medical and social research’ (Bailey 1978: 384). Informedconsentimpliesthatsubjectsaremadeadequatelyawareofthetypeofinformationyouwantfromthem,whytheinformationisbeingsought,whatpurposeitwillbeputto,howtheyareexpectedtoparticipatein thestudy,andhowitwilldirectlyor indirectlyaffect them. It is important that theconsentshouldalsobevoluntaryandwithoutpressureofanykind.SchinkeandGilchristwrite:
Under standards set by the National Commission for the Protection of Human Subjects, allinformed-consent procedures must meet three criteria: participants must be competent to giveconsent; sufficient informationmust be provided to allow for a reasoned decision; and consentmustbevoluntaryanduncoerced.(1993:83)
Competency,accordingtoSchinkeandGilchrist,‘isconcernedwiththelegalandmentalcapacitiesofparticipantstogivepermission’(1993:83).Forexample,someveryoldpeople,thosesufferingfromconditions that exclude them frommaking informed decisions, people in crisis, people who cannotspeakthelanguageinwhichtheresearchisbeingcarriedout,peoplewhoaredependentuponyouforaserviceandchildrenarenotconsideredtobecompetent.
Providingincentives
Is it ethical to provide incentives to respondents to share information with you? Some researchersprovide incentives to participants for their participation in a study, feeling this to be quite proper asparticipantsaregivingtheirtime.Othersthinkthattheofferingofinducementsisunethical.In the author’s experience most people do not participate in a study because of incentives, but
because they realise the importanceof the study.Therefore,givinga smallgift afterhavingobtainedyourinformation,asatokenofappreciation,isintheauthor’sopinionnotunethical.However,givingapresentbeforedatacollectionisunethical.
Seekingsensitiveinformation
Information sought can pose an ethical dilemma in research. Certain types of information can beregardedassensitiveorconfidentialbysomepeopleand thusan invasionofprivacy.Askingfor thisinformationmayupsetorembarrassa respondent.However, ifyoudonotask for the information, itmaynotbepossibletopursueyourinterestintheareaandcontributetotheexistingbodyofknowledge.For most people, questions on sexual behaviour, drug use and shoplifting are intrusive. Even
questionsonmaritalstatus,incomeandagemaybeconsideredtobeaninvasionofprivacybysome.Incollectingdatayouneedtobecarefulaboutthesensitivitiesofyourrespondents.
Thedilemmayoufaceasaresearcheriswhetheryoushouldasksensitiveandintrusivequestions.Intheauthor’sopinionitisnotunethicaltoasksuchquestionsprovidedthatyouclearlyandfranklytellyourrespondentsthetypeofinformationyouaregoingtoask,andgivethemsufficienttimetodecideiftheywanttosharetheinformationwithyou,withoutanymajorinducement.
Thepossibilityofcausingharmtoparticipants
Istheresearchgoingtoharmparticipantsinanyway?Harmincludes:
notonlyhazardousmedicalexperimentsbutalsoanysocialresearchthatmightinvolvesuchthingsasdiscomfort,anxiety,harassment,invasionofprivacy,ordemeaningordehumanisingprocedures.(Bailey1978:384)
Whenyoucollectdatafromrespondentsorinvolvesubjectsinanexperiment,youneedtoexaminecarefullywhethertheirinvolvementislikelytoharmtheminanyway.Ifitis,youmustmakesurethattherisk isminimal.Minimumriskmeans that theextentofharmordiscomfort in theresearch isnotgreater than that ordinarily encountered in daily life. It is unethical if theway you seek informationcreatesanxietyorharassment,andifyouthinkitmayhappen,youneedtotakestepstopreventthis.
Maintainingconfidentiality
Sharing information about a respondent with others for purposes other than research is unethical.Sometimes you need to identify your study population to put your findings into context. In such asituation you need to make sure that at least the information provided by respondents is keptanonymous. It is unethical to identify an individual respondent and the information provided byhim/her.Therefore,youneedtoensurethataftertheinformationhasbeencollected,itssourcecannotbeidentified.Incertaintypesofstudyyoumightneedtovisitrespondentsrepeatedly,inwhichcaseyouwillhave to identify themuntil thecompletionofyourvisits. Insuchsituationsyouneed tobeextracarefulthatothersdonothaveaccesstotheinformation.Itisunethicaltobenegligentinnotprotectingtheconfidentialityandanonymityoftheinformationgatheredfromyourrespondents.Ifyouaredoingresearch for someone else, you need tomake sure that confidentiality ismaintained by this party aswell.
Ethicalissuestoconsiderrelatingtotheresearcher
Avoidingbias
Bias on the part of the researcher is unethical. Bias is different from subjectivity. Subjectivity, asmentionedearlier,isrelatedtoyoureducationalbackground,trainingandcompetenceinresearch,andyourphilosophicalperspective.Biasisadeliberateattempteithertohidewhatyouhavefoundinyourstudy, or to highlight something disproportionately to its true existence. It is absolutely unethical tointroducebiasintoaresearchactivity.Ifyouareunabletocontrolyourbias,youshouldnotbeengagingintheresearch.Remember,itisthebiasthatisunethicalandnotthesubjectivity.
Provisionordeprivationofatreatment
Boththeprovisionanddeprivationofatreatmentmayposeanethicaldilemmaforyouasaresearcher.Whentestinganinterventionoratreatment,aresearcherusuallyadoptsacontrolexperimentdesign.Insuchstudies,isitethicaltoprovideastudypopulationwithaninterventionortreatmentthathasnotyetbeenconclusivelyproveneffectiveorbeneficial?But ifyoudonot testa treatment/intervention,howcan you prove or disprove its effectiveness or benefits? On the other hand, you are providing anintervention thatmay not be effective. Is this ethical? Is it ethical to deprive the control group of atreatmentevenifitmayprovetobeonlyslightlyeffective?Andbeyondtheissueofcontrolgroups,isitethicaltodeprivepeoplewhoarestrugglingforlifeofthepossiblebenefit,howeversmall,whichmaybederivedfromadrugthatisonlyundertrial?Asaresearcheryouneedtobeawareoftheseethicalissues.Thereareargumentsandcounter-argumentsabouttheseissues.However,itisusuallyacceptedthatdeprivationofatrialtreatmenttoacontrolgroupisnotunethicalas,intheabsenceofthis,astudycan never establish the effectiveness of a treatment which may deprive many others of its possiblebenefits. This deprivation of the possible benefits, on the other hand, is considered by some asunethical.There are no simple answers to these dilemmas. Ensuring informed consent, ‘minimum risk’ and
frankdiscussionastotheimplicationsofparticipationinthestudymayhelptoresolvesomeoftheseethicalissues.
Usinginappropriateresearchmethodology
A researcher has an obligation to use appropriate methodology, within his/her knowledge base, inconducting a study. It is unethical to use deliberately a method or procedure you know to beinappropriate to prove or disprove something that youwant to, such as by selecting a highly biasedsample,usinganinvalidinstrumentorbydrawingwrongconclusions.
Incorrectreporting
Toreportthefindingsinawaythatchangesorslantsthemtoserveyourownorsomeoneelse’sinterestis unethical. Correct and unbiased reporting of the findings are important characteristics of ethicalresearchpractice.
Inappropriateuseoftheinformation
Howwilltheinformationobtainedfromrespondentsbeusedbytheresearcher?Theuseofinformationinawaythatdirectlyorindirectlyaffectsrespondentsadverselyisunethical.Caninformationbeusedadversely to affect the study population? If so, how can the study population be protected? As aresearcheryouneedtoconsiderandresolvetheseissues.Sometimesitispossibletoharmindividualsinthe process of achieving benefits for organisations. An example would be a study to examine thefeasibility of restructuring an organisation. Restructuring may be beneficial to the organisation as awhole butmay be harmful to some individuals. Should you ask respondents for information that islikelytobeusedagainstthem?Ifyoudo,theinformationmaybeusedagainstthem,andifyoudonot,theorganisationmaynot be able toderive thebenefits of restructuring. In the author’s opinion, it isethicaltoaskquestionsprovidedyoutellrespondentsofthepotentialuseoftheinformation,including
the possibility of its being used against some of them, and you let them decide if they want toparticipate.Somemayparticipateforthebettermentoftheorganisationeventhoughitmayharmthemandothersmaydecideagainstit.However,toidentifyeitherofthemisunethicalinresearch.
Ethicalissuesregardingthesponsoringorganisation
Restrictionsimposedbythesponsoringorganisation
Mostresearchinthesocialsciencesiscarriedoutusingfundsprovidedbysponsoringorganisationsfora specific purpose. The fundsmay be given to develop a programme or evaluate it; to examine itseffectivenessandefficiency;tostudytheimpactofapolicy;totestaproduct;tostudythebehaviourofagrouporcommunity;ortostudyaphenomenon,issueorattitude.Sometimestheremaybedirectorindirectcontrolsexercisedbysponsoringorganisations.Theymayselectthemethodology,prohibitthepublicationof‘whatwasfound’orimposeotherrestrictionsontheresearchthatmaystandinthewayof obtaining and disseminating accurate information. Both the imposition and acceptance of thesecontrols and restrictions are unethical, as they constitute interference and could amount to thesponsoringorganisationtailoringresearchfindingstomeetitsvestedinterests.
Themisuseofinformation
How is the sponsoring body going to use the information? How is this likely to affect the studypopulation?Sometimessponsoringorganisationsuseresearchasapretextforobtainingmanagement’sagenda. It is unethical to let your research be used as a reason for justifyingmanagement decisionswhen the research findings do not support them.However, it is recognised that itmay be extremelydifficultorevenimpossibleforaresearchertopreventthisfromhappening.
SummaryThischapterisdesignedtomakeyouawareoftheethicalissuestobeconsideredwhenconductingresearch.Theethicalissuestobeconsideredarethesameinbothquantitativeandqualitativeresearch.Howyouresolvethemdependsuponyou,andtheconditionsunderwhichyouareworking.Beingethicalmeansadheringtothecodeofconductthathasevolvedovertheyearsforanacceptableprofessionalpractice.Any
deviationfromthiscodeofconductisconsideredasunethicalandthegreaterthedeviation,themoreseriousthebreach.Formostprofessionsethicalcodesinresearchareanintegralpartoftheiroverallethics,thoughsomeresearchbodieshaveevolvedtheirowncodes.Ethicalissuesinresearchcanbelookedatastheyrelatetoresearchparticipants,researchersandsponsoringorganisations.With
regardtoresearchparticipants,thefollowingareascouldposeethicalissuesifnotdealtwithproperly:collectinginformation;seekingconsent; providing incentives; seeking sensitive information; the possibility of causing harm to participants; and maintainingconfidentiality.It is important toexaminetheseareasthoroughlyforanyunethicalpractice.Withregardto theresearcher,areasofethicalconcernincludethefollowing:introducingbias;providinganddeprivingindividualsoftreatment;usingunacceptableresearchmethodology; inaccurate reporting; and the inappropriate use of information. Ethical considerations in relation to sponsoringorganisations concern restrictions imposed on research designs and the possible use of findings.As a newcomer to research youshouldbeawareofwhatconstitutesunethicalpracticeandbeabletoputappropriatestrategiesinplacetodealwithanyharmthatmaydonetoanystakeholder.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Findacopyofyouruniversity’sordepartment’scodeofethicsforresearch(orexamplesofcodesofconductforyourchosenprofession).Canyouidentifyanyareasofresearchorapproachesthatmightcomeintoconflictwiththeseguidelines?Somemightsuggestthataskingforanykindofinformationfromanindividualisunethicalasitisaninvasionofhis/herprivacy.Considerhowyoumightargueforandagainstthissuggestion.Ethicalissuesmayariseatanypointintheresearchprocess.Reflectingontheprinciplesraisedinthischapter,makealistofethicalissuesthatyouthinkshouldbeconsideredateachstepintheeight-stepmodel.Imagineyouareplanningtoundertakeahypotheticalresearchstudyinanareaofinteresttoyou.Identifythevariousstakeholdergroupsandlistthepossibleethicalconcernsyouneedtobeawareoffromtheperspectiveofeachoneofthegroups.
STEPVIIProcessingandDisplayingData
Thisoperationalstepincludestwochapters:
Chapter15:ProcessingdataChapter16:Displayingdata
CHAPTER15ProcessingData
Inthischapteryouwilllearnabout:
MethodsforprocessingdatainquantitativestudiesHowtoeditdataandpreparedataforcodingHowtocodedataHowtocodequalitativedatainquantitativestudiesMethodsforprocessingdatainqualitativestudiesAnalysingdatainqualitativeandquantitativestudiesTheroleofcomputersindataanalysisTheroleofstatisticsinresearch
Keywords: analysis, closed questions, code book, coding, concepts, contentanalysis, cross-tabulation, data displaying, data processing, editing, frame ofanalysis,frequencydistribution,multipleresponses,open-endedquestions,pre-test.
Ifyouwereactuallydoingaresearchstudy,youwouldbynowhavereachedastagewhereyouhaveeitherextractedorcollectedtherequiredinformation.Thenextstepiswhattodowiththisinformation.Howdoyoufindtheanswerstoyourresearchquestions?Howdoyoumakesenseoftheinformationcollected?Howdoyouproveordisproveyourhypothesisifyouhadone?Howshouldtheinformationbeanalysedtoachievetheobjectivesofyourstudy?Toanswerthesequestionsyouneedtosubjectyourdatatoanumberofproceduresthatconstitutethecoreofdataprocessing(Figure15.1).
FIGURE15.1Stepsindataprocessing
These procedures are the samewhether your study is quantitative or qualitative, butwhat you dowithineachprocedureisdifferent.Forbothtypesofstudyyouneedtovisualisehowyouaregoingtopresentyour findings toyour readership in lightof itsbackgroundand thepurposeof thestudy.Youneedtodecidewhattypeofanalysiswouldbeappropriateforthereadersofyourreport.Itisinlightofthepurposeofyourstudyandyourimpressionaboutthelevelofunderstandingofyourreadershipthatyou decide the type of analysis you should undertake. For example, there is no point in doing asophisticated statistical analysis if your readers are not familiar with statistical procedures. Inquantitative research themain emphasis in data analysis is to decide how you are going to analyseinformation obtained in response to each question that you asked of your respondents. In qualitativeresearchthefocusisonwhatshouldbethebasisofanalysisof theinformationobtained; that is, is itcontents,discourse,narrativeoreventanalysis?Becauseofthedifferenttechniquesusedinprocessingdatainquantitativeandqualitativeresearch,thischapterisdividedintotwoparts.PartOnedealswithdataprocessinginquantitativestudiesandPartTwowithqualitative.
Partone:Dataprocessinginquantitativestudies
Editing
Irrespective of themethod of data collection, the information collected is called raw data or simplydata. The first step in processing your data is to ensure that the data is ‘clean’ – that is, free frominconsistenciesandincompleteness.Thisprocessof‘cleaning’iscalledediting.Editingconsistsofscrutinisingthecompletedresearchinstrumentstoidentifyandminimise,asfaras
possible, errors, incompleteness, misclassification and gaps in the information obtained from therespondents.Sometimeseventhebestinvestigatorscan:
forgettoaskaquestion;forgettorecordaresponse;wronglyclassifyaresponse;writeonlyhalfaresponse;writeillegibly.
Inthecaseofaquestionnaire,similarproblemscancropup.Theseproblemstoagreatextentcanbereduced simply by (1) checking the contents for completeness, and (2) checking the responses forinternalconsistency.Thewayyoucheckthecontentsforcompletenessdependsuponthewaythedatahasbeencollected.
Inthecaseofaninterview,justcheckingtheinterviewschedulefortheaboveproblemsmayimprovethequalityofthedata.Itisgoodpracticeforaninterviewertotakeafewmomentstoperuseresponsesforpossibleincompletenessandinconsistencies.Inthecaseofaquestionnaire,again,justbycarefullychecking theresponsessomeof theproblemsmaybereduced.Thereareseveralwaysofminimisingsuchproblems:
Byinference–Certainquestionsinaresearchinstrumentmayberelatedtooneanotheranditmightbepossibletofindouttheanswertoonequestionfromtheanswertoanother.Ofcourse,youmustbecarefulaboutmakingsuchinferencesoryoumayintroducenewerrorsintothedata.Byrecall–Ifthedataiscollectedbymeansofinterviews,sometimesitmightbepossiblefortheinterviewertorecallarespondent’sanswers.Again,youmustbeextremelycareful.Bygoingbacktotherespondent–Ifthedatahasbeencollectedbymeansofinterviewsorthequestionnairescontainsomeidentifyinginformation,itispossibletovisitorphonearespondenttoconfirmorascertainananswer.Thisis,ofcourse,expensiveandtimeconsuming.
Therearetwowaysofeditingthedata:
1. examinealltheanswerstoonequestionorvariableatatime;2. examinealltheresponsesgiventoallthequestionsbyonerespondentatatime.
Theauthorprefersthesecondmethodasitprovidesatotalpictureoftheresponses,whichalsohelpsyoutoassesstheirinternalconsistency.
Coding
Having‘cleaned’thedata,thenextstepistocodeit.Themethodofcodingislargelydictatedbytwo
considerations:
1. thewayavariablehasbeenmeasured(measurementscale)inyourresearchinstrument(e.g.ifaresponsetoaquestionisdescriptive,categoricalorquantitative);
2. thewayyouwanttocommunicatethefindingsaboutavariabletoyourreaders.
Forcoding,thefirstlevelofdistinctioniswhetherasetofdataisqualitativeorquantitativeinnature.For qualitative data a further distinction is whether the information is descriptive in nature (e.g. adescription of a service to a community, a case history) or is generated through discrete qualitativecategories. For example, the following information about a respondent is in discrete qualitativecategories: income – above average, average, below average; gender – male, female; religion –Christian,Hindu,Muslim,Buddhist,etc.;orattitudetowardsanissue–stronglyfavourable,favourable,uncertain,unfavourable,stronglyunfavourable.Eachofthesevariablesismeasuredeitheronanominalscaleoranordinalscale.Someofthemcouldalsohavebeenmeasuredonaratioscaleoranintervalscale.Forexample,incomecanbemeasuredindollars(ratioscale),oranattitudetowardsanissuecanbemeasuredonan intervalor a ratio scale.Thewayyouproceedwith thecodingdependsupon themeasurement scale used in themeasurement of a variable andwhether a question is open-ended orclosed.Inaddition, the typesofstatisticalprocedures thatcanbeapplied toasetof information toa large
extent depend upon the measurement scale on which a variable was measured in the researchinstrument. For example, you can find out different statistical descriptors such as mean, mode andmedianifincomeismeasuredonaratioscale,butnotifitismeasuredonanordinaloranominalscale.It is extremely important to understand that theway you are able to analyse a set of information isdependentuponthemeasurementscaleusedin theresearchinstrumentformeasuringavariable.It istherefore important to visualise – particularly at the planning stage when constructing the researchinstrument–thewayyouaregoingtocommunicateyourfindings.Howyoucananalyseinformationobtainedinresponsetoaquestiondependsuponhowaquestion
wasasked,andhowarespondentansweredit.Inotherwords,itdependsuponthemeasurementscaleonwhich a response can bemeasured/classified. If you study answers given by your respondents inreplytoaquestion,youwillrealisethatalmostallresponsescanbeclassifiedintooneofthefollowingthreecategories:
1. quantitativeresponses;2. categoricalresponses(whichmaybequantitativeorqualitative);3. descriptiveresponses(whichareinvariablyqualitative–keepinmindthatthisisqualitativedata
collectedaspartofquantitativeresearchandnotthequalitativeresearch).
Forthepurposeofanalysis,quantitativeandcategoricalresponsesneedtobedealtwithdifferentlyfrom descriptive ones. Both quantitative and categorical information go through a process that isprimarily aimed at transforming the information into numerical values, called codes, so that theinformation can be easily analysed, eithermanually or by computers.On the other hand, descriptiveinformationfirstgoesthroughaprocesscalledcontentanalysis,wherebyyouidentifythemainthemesthatemerge fromthedescriptionsgivenby respondents inanswer toquestions.Having identified themainthemes,therearethreewaysthatyoucandealwiththem:(1)youcanexamineverbatimresponsesandintegratethemwiththetextofyourreporttoeithersupportorcontradictyourargument;(2)youcanassignacodetoeachthemeandcounthowfrequentlyeachhasoccurred;and(3)youcancombineboth
methodstocommunicateyourfindings.Thisisyourchoice,anditisbasedonyourimpressionofthepreferenceofyourreaders.For coding quantitative and qualitative data in quantitative studies you need to go through the
followingsteps:
StepIdevelopingacodebook;StepIIpre-testingthecodebook;StepIIIcodingthedata;StepIVverifyingthecodeddata.
StepI:Developingacodebook
A code book provides a set of rules for assigning numerical values to answers obtained fromrespondents.Letustakeanexample.Figure15.2listssomequestionstakenfromaquestionnaireusedin a survey conducted by the author to ascertain the impact of occupational redeployment on anindividual.Thequestions selected shouldbe sufficient to serve as a prototype for developing a codebook,astheycoverthevariousissuesinvolvedintheprocess.
FIGURE15.2Exampleofquestionsfromasurvey
There are two formats fordata entry: ‘fixed’ and ‘free’. In this chapterwewill beusing the fixed
formattoillustratehowtodevelopacodebook.Thefixedformatstipulatesthatapieceofinformationobtainedfromarespondent isenteredinaspecificcolumn.Eachcolumnhasanumberandthe‘Col.no.’inthecodebookreferstothecolumninwhichaspecifictypeofinformationistobeentered.Theinformationaboutanindividualisthusenteredinarow(s)comprisingthesecolumns.Forabeginneritisimportanttounderstandthestructureofacodebook(Table15.1),whichisbased
ontheresponsesgiventothequestionslistedinFigure15.2.InTable15.1, column 1 refers to the columns inwhich a particular piece of information is to be
entered. Allocation of columns in a fixed format is extremely important because, when youwrite aprogram,youneedtospecifythecolumninwhichaparticularpieceofinformationisenteredsothatthecomputercanperformtherequiredprocedures.Column 2 identifies the question number in the research instrument for which the information is
beingcoded.Thisisprimarilytoidentifycodingwiththequestionnumberintheinstrument.Column3referstothenameofthevariable.Eachvariableinaprogramisgivenauniquenameso
thattheprogramcancarryouttherequestedstatisticalprocedures.Usuallytherearerestrictionsonthewayyoucannameavariable(e.g.thenumberofcharactersyoucanusetonameavariableandwhetheryouusethealphabetornumerals).Youneedtocheckyourprogramforthis.Itisadvisabletonameavariableinsuchawaythatyoucaneasilyrecogniseitfromitsname.Column4liststheresponsestothevariousquestions.Developingaresponsepatternforthequestions
is themost important, difficult and time-consuming part of developing a code book. The degree ofdifficulty in developing a response pattern differs with the types of questions in your researchinstrument (open ended or closed). If a question is closed, the response pattern has already beendeveloped as part of the instrument construction and all you need to do at this stage is to assign anumericalvaluetoeachresponsecategory.Intermsofanalysis,thisisoneofthemainadvantagesofclosed questions. If a closed question includes ‘other’ as one of the response categories, toaccommodateanyresponsethatyoumaynothavelistedwhendevelopingtheinstrument,youshouldanalysetheresponsesandassignthemtonon-overlappingcategoriesinthesamewayasyouwoulddofor open-ended questions.Add these to the already developed response categories and assign each anumericalvalue.If the number of responses to a question is less thannine, youneedonly one column to code the
responses,and if it ismore thanninebut less than99,youneed twocolumns (column1 in thecodebook). But if a question asks respondents to give more than one response, the number of columnsassigned should be in accordancewith the number of responses to be coded. If there are, say, eightpossibleresponsestoaparticularquestionandarespondentisaskedtogivethreeresponses,youneedthreecolumnstocodetheresponsestothequestion.Letusassumethereare12possibleresponsestoaquestion.Tocodeeachresponseyouneedtwocolumnsand,therefore,tocodethreeresponsesyouneedsixcolumns.Thecodingofopen-endedquestionsismoredifficult.Codingofopen-endedquestionsrequiresthe
responsecategories tobedeveloped first throughaprocesscalledcontentanalysis.Oneof theeasierways of analysing open-ended questions is to select a number of interview schedules/questionnairesrandomlyfromthetotalcompletedinterviewschedulesorquestionnairesreceived.Thenselectanopen-endedquestionfromoneoftheseschedulesorquestionnairesandwritedowntheresponse(s)onasheetof paper. If the person has givenmore than one response,write them separately on the same sheet.Similarly,fromthesamequestionnaire/scheduleselectanotheropen-endedquestionandwritedowntheresponsesgivenona separate sheet. In the samewayyoucanselectotheropen-endedquestionsandwrite down the response(s). Remember that the response to each question should be written on aseparatesheet.Nowselectanotherquestionnaire/interviewscheduleandgothroughthesameprocess,addingresponse(s)givenforthesamequestiononthesheetforthatquestion.Continuetheprocessuntil
youfeel that theresponsesarebeingrepeatedandyouaregettingnoorveryfewnewones– that is,whenyouhavereachedasaturationpoint.
TABLE15.1Anexampleofacodebook
Now,onebyone,examinetheresponsestoeachquestiontoascertainthesimilaritiesanddifferences.If two ormore responses are similar inmeaning though not necessarily in language, try to combinethemunderonecategory.Giveanametothecategorythatisdescriptiveoftheresponses.Remember,whenyou code thedata you code categories,not responses per se. It is advisable towrite down thedifferentresponsesundereachcategoryinthecodebooksothat,whilecoding,youknowthetypeofresponsesyouhavegroupedunderacategory.Indevelopingthesecategoriestherearethreeimportantconsiderations:
1. Thecategoriesshouldbemutuallyexclusive.Developnon-overlappingcategories.Aresponseshouldnotbeabletobeplacedwithintwocategories.
2. Thecategoriesshouldbeexhaustive;thatis,almosteveryresponseshouldbeabletobeplacedwithinoneofthecategories.Iftoomanyresponsescannotbesocategorised,itisanindicationofineffectivecategorisation.Insuchasituationyoushouldexamineyourcategoriesagain.
3. Theuseofthe‘other’category,effectivelya‘wastebasket’forthoseoddresponsesthatcannotbeputintoanycategory,mustbekepttotheabsoluteminimumbecause,asmentioned,itreflectsthefailureoftheclassificationsystem.Thiscategoryshouldnotincludemorethan5percentofthetotalresponsesandshouldnotcontainanymoreresponsesthananyothercategory.
Column5 lists theactualcodesof thecodebook thatyoudecide toassign toa response.Youcanassignanynumericalvaluetoanyresponsesolongasyoudonotrepeatitforanotherresponsewithin
the same question. Two responses to questions are commonly repeated: ‘not applicable’ and ‘noresponse’.Youshouldselectanumberthatcanbeusedfortheseresponsesforallormostquestions.Forexample, responses such as ‘not applicable’ and ‘no response’ could be given a code of 8 and 9respectively, even though the responses to aquestionmaybe limited toonly2or3. Inotherwords,supposeyouwanttocodethegenderofarespondentandyouhavedecidedtocodefemale=1andmale=2.For‘noresponse’, insteadofassigningacodeof3,assignacodeof9.Thissuggestionhelps inrememberingcodes,whichwillhelptoincreaseyourspeedincoding.Toexplainhowtocode,letustakethequestionslistedintheexampleinFigure15.2.Wewilltake
eachquestiononebyonetodetailtheprocess.
Question1(a)
Yourcurrentageincompletedyears:______
Thisisanopen-endedquantitativequestion.Inquestionslikethisitisimportanttodeterminetherangeofresponses–therespondentwiththelowestandtherespondentwiththehighestage.Todothis,gothroughanumberofquestionnaires/interviewschedules.Oncetherangeisestablished,divideitintoanumberofcategories.Thecategoriesdevelopedaredependentuponanumberofconsiderationssuchasthepurposeofanalysis,thewayyouwanttocommunicatethefindingsofyourstudyandwhetherthefindingsaregoingtobecomparedwiththoseofanotherstudy.Letusassumethattherangeinthestudyis23to49yearsandassumethatyoudevelopthefollowingcategoriestosuityourpurpose:20–24,25–29,30–34,35–39,40–44and45–49.Ifyourrangeiscorrectyoushouldneednoothercategories.Letusassume that you decide to code 20–24= 1, 25–29= 2, 30–34= 3, and so on.To accommodate ‘noresponse’youdecide toassigna codeof9.Letus assumeyoudecided to code the responses to thisquestionincolumn5ofthecodesheet.
Question1(c)
Yourmaritalstatus:(Pleasetick)Currentlymarried________Livinginadefactorelationship____Separated______________Divorced_______________Nevermarried__________
This is a closed categorical question. That is, the response pattern is already provided. In thesesituationsyoujustneedtoassignanumericalvaluetoeachcategory.Forexample,youmaydecidetocode‘currentlymarried’=1,‘livinginadefactorelationship’=2,‘separated’=3,‘divorced’=4and‘nevermarried’=5.Youmayadd‘noresponse’asanothercategoryandassignitwithacodeof9.Theresponsetothisquestioniscodedincolumn6ofthecodesheet.
Question2(b)
If tertiary/university, please specify the level achieved and the area of study. (Please specify allpostgraduatequalifications.)
In this question a respondent is asked to indicate the area in which s/he has achieved a tertiaryqualification.Thequestionasksfortwoaspects:(1)levelofachievement,whichiscategorical;and(2)areaofstudy,whichisopenended.Also,apersonmayhavemorethanonequalificationwhichmakesitamultipleresponsequestion.Insuchquestionsbothaspectsofthequestionaretobecoded.Inthiscase,this means the level of achievement (e.g. associate diploma, diploma) and the area of study (e.g.engineering,accounting).Whencodingmultipleresponses,decideonthemaximumpossiblenumberofresponsestobecoded.Letusassumeyoucodeamaximumnumberofthreelevelsoftertiaryeducation.(This would depend upon the maximum number of levels of achievement identified by the studypopulation.)Firstly,codethelevelsofachievementTEDU(TEDU:T=tertiaryandEDU=education; thenamingofthevariable–‘levelofachievement’–inthismannerisdoneforeasyidentification)andthentheareaofStudy,STUDY(thevariablenamegiventothe‘areaofstudy’=STUDY).Intheaboveexample,letusassumethatyoudecidedtocodethreelevelsofachievement.Todistinguishthemfromeachotherwe call the first level TEDU1, the second TEDU2 and the third TEDU3, and decide to code them incolumns, 7, 8 and 9 respectively. Similarly, the names given to the three areas of STUDY1, STUDY2,STUDY3andwedecidetocodethemincolumns10–11,12–13and14–15.Thecodes(01to23)assignedto different qualifications are listed in the code book. If a respondent has only one qualification, thequestion of second and third qualification is not applicable and you need to decide a code for ‘notapplicable’.Assumeyouassignedacodeof88.‘Noresponse’wouldthenbeassignedacodeof99forthisquestion.
Question11
What,inyouropinion,arethemaindifferencesbetweenyourjobspriortoandafterredeployment?
Thisisanopen-endedquestion.Tocodethisyouneedtogothroughtheprocessofcontentanalysisasexplainedearlier.Withinthescopeofthischapteritisnotpossibletoexplainthedetails,butresponsecategoriesthathavebeenlistedarebasedupontheresponsesgivenby109respondentstothesurveyonoccupational redeployment. In coding questions like this, on the one hand you need to keep thevariation in the respondents’ answers and, on the other, youwant to break them up intomeaningfulcategories to identify the commonalities. Because this question is asking respondents to identify thedifferencesbetweentheirjobsbeforeandafterredeployment,foreasyidentificationletusassumethisvariablewasnamedDIFWK(DIF=differenceandWK=work).Responses to thisquestionare listed inFigure15.3.Theseresponseshavebeenselectedatrandomfromthequestionnairesreturned.Acloseexaminationoftheseresponsesrevealsthatanumberofthemesarecommon,forexample:
‘learning new skills in the new job’; ‘challenging tasks are missing from the new position’; ‘moresecureinthepresentjob’;‘moreinteractioninthepresentjob’;‘lessresponsibility’;‘morevariety’;‘nodifference’; ‘more satisfying’.There aremany similar themes that hold for both the before and afterjobs.Therefore,wedevelopedthesethemesfor‘currentjob’and‘previousjob’.Oneof themaindifferencesbetweenqualitativeandquantitative research is thewayresponsesare
used in the report. In qualitative research the responses are normally used either verbatim or areorganisedundercertainthemesandtheactualresponsesareprovidedtosubstantiatethem.Inquantitativeresearch theresponsesareexamined,commonthemesare identified, the themesare
named(orcategoriesaredeveloped)andtheresponsesgivenbyrespondentsareclassifiedunderthese
themes.Thedatathencanalsobeanalysedtodeterminethefrequencyofthethemesifsodesired.Itisalsopossibletoanalysethethemesinrelationtosomeothercharacteristicssuchasage,educationandincomeofthestudypopulation.
FIGURE15.3Someselectedresponsestotheopen-endedquestion(no.11)inFigure15.2
The codebook lists the themesdevelopedon the basis of responses given.Asyou can see,manycategories may result. The author’s advice is not to worry about this as categories can always becombinedlaterifrequired.Thereverseisimpossibleunlessyougobacktotherawdata.Letusassumeyouwanttocodeuptofiveresponsestothisquestionandthatyouhavedecidedto
namethesefivevariablesasDIFWK1DIFWK2,DIFWK3,DIFWK4andDIFWK5.Letusalsoassumethatyouhavecodedthemincolumns16–17,18–19,20–21,22–23and24–25respectively.
Question12
Wewouldliketoknowyourlevelofsatisfactionwiththetwojobsbeforeandafterredeploymentwithrespecttothefollowingaspectsofyourjob.Pleaseratethemonafive-pointscaleusingthefollowingguide:
5=extremelysatisfied,4=verysatisfied,3=satisfied,2=dissatisfied,1=extremelydissatisfied
This isahighlystructuredquestionaskingrespondents tocompareona five-pointordinalscale theirlevelofsatisfactionwithvariousareasof their jobbeforeandafterredeployment.Aswearegaugingthelevelofsatisfactionbeforeandafterredeployment,respondentsareexpectedtogivetworesponsesto each area. In this example let us assume you have used the name JOBSTA for job status afterredeployment(JOB=job,ST=statusandA=afterredeployment)andJOBSTB forbefore redeployment(JOB=job,ST=statusandB=beforeredeployment).Similarly,forthesecondarea,jobsatisfaction,youhavedecidedthatthevariablename,JOBSATA(JOB=job,SAT=satisfactionandA=after),willstandforthe level of job satisfaction after redeployment and JOBSATB will stand for the level beforeredeployment.Othervariablenameshavebeensimilarlyassigned.Inthisexamplethevariable,JOBSTA,isenteredincolumn26,JOBSTAincolumn27,andsoon.
StepII:Pre-testingthecodebook
Onceacodebookisdesigned,itisimportanttopre-testitforanyproblemsbeforeyoucodeyourdata.A pre-test involves selecting a few questionnaires/interview schedules and actually coding theresponsestoascertainanyproblemsincoding.Itispossiblethatyoumaynothaveprovidedforsomeresponsesandthereforewillbeunabletocodethem.Changeyourcodebook,ifyouneedto,inlightofthepre-test.
StepIII:Codingthedata
Onceyourcodebookisfinalised,thenextstepistocodetherawdata.Therearethreewaysofdoingthis:
1. codingonthequestionnaires/interviewscheduleitself,ifspaceforcodingwasprovidedatthetimeofconstructingtheresearchinstrument;
2. codingonseparatecodesheetsthatareavailableforpurchase;3. codingdirectlyintothecomputerusingaprogramsuchasSPSSx,SAS.
Toexplaintheprocessofcodingletustakethesamequestionsthatwereusedindevelopingthecodebook.We select three questionnaires at random from a total of 109 respondents (Figures 15.4, 15.5,15.6).Usingthecodebookasaguide,wecodetheinformationfromthesesheetsontothecodingsheet(Figure15.7).Letusexaminethecodingprocessbytakingrespondent3(Figure15.4).
Respondent3Thetotalnumberofrespondentsismorethan99andthisisthethirdquestionnaire,so003wasgivenastheidentificationnumberwhichiscodedincolumns1–3(Figure15.7).Becauseitisthefirstrecordforthisrespondent,1wascodedincolumn4.Thisrespondentis49yearsofageandfallsinthecategory45–49,whichwascodedas6.Astheinformationonageisenteredincolumn5,6wascodedinthiscolumnofthecodesheet.Themaritalstatusofthispersonis‘divorced’,hence4wascodedincolumn6.ThispersonhasaBachelorsdegree in librarianship.Thecodechosen foraBachelorsdegree is3,whichwasenteredincolumn7.Threetertiaryqualificationshavebeenprovidedfor,andasthispersondoesnot have anyother qualifications,TEDU2TEDU3 are not applicable, and therefore a codeof 8 isenteredincolumns8and9.Thisperson’sBachelorsdegreeisinlibrarianshipforwhichcode09was
assignedandenteredincolumns10–11.Sincethereisonlyonequalification,STUDY2andSTUDY3arenotapplicable;therefore,acodeof88wasenteredincolumns12–13and14–15.Thispersonhasgivenanumber of responses to question no. 11 (DIFWK),which asks respondents to list themaindifferencesbetweentheirjobsbeforeandafterredeployment.Incodingsuchquestionsmuchcautionisrequired.Examine theresponsesnamedDIFWK1,DIFWK2,DIFWK3,DIFWK4,DIFWK5, to identify thecodes that
canbeassigned.Acodeof22(nowdealwithpublic)wasassignedtooneoftheresponses,whichweenter in columns16–17.The seconddifference,DIFWK2,was assigneda codeof69 (totallydifferentskillrequired),whichiscodedincolumns18–19.
DIFWK3 was assigned a code of 77 (current job more structure) and coded in columns 20–21.Similarly, the fourth (DIFWK4) and the fifth (DIFWK5) difference in the jobs before and afterredeploymentarecodedas78(nowpartof the teaminsteadof independentworker)and38(hours–nowfulltime),whichareenteredincolumns22–23and24–25respectively.Question12isextremelysimpletocode.Eachareaofajobhastwocolumns,oneforbeforeandtheotherforafter.Jobstatus(JOBST)isdividedintotwovariables,JOBSTAforarespondent’slevelofsatisfactionafterredeploymentandJOBSTBforhis/herlevelbeforeredeployment.JOBSTAisenteredincolumn26andJOBSTBincolumn27.For JOBSTA the code, 5 (asmarkedby the respondent), is entered in column26 and the code forJOBSTB,4, isenteredincolumn27.Otherareasofthejobbeforeandafterredeploymentaresimilarlycoded.Theothertwoexamplesarecodedinthesamemanner.ThecodeddataisshowninFigure15.7. In
theprocessofcodingyoumightfindsomeresponsesthatdonotfityourpredeterminedcategories.Ifso,assignthemacodeandaddthesetoyourcodebook.
FIGURE15.4Somequestionsfromasurvey–respondent3
FIGURE15.5Somequestionsfromasurvey–respondent59
FIGURE15.6Somequestionsfromasurvey–respondent81
FIGURE15.7Anexampleofcodeddataonacodesheet
StepIV:Verifyingthecodeddata
Oncethedataiscoded,selectafewresearchinstrumentsatrandomandrecordtheresponsestoidentifyanydiscrepanciesincoding.Continuetoverifycodinguntilyouaresurethattherearenodiscrepancies.Iftherearediscrepancies,re-examinethecoding.
Developingaframeofanalysis
Althoughaframeworkofanalysisneedstoevolvecontinuouslywhilewritingyourreport,itisdesirabletobroadlydevelopitbeforeanalysingthedata.Aframeofanalysisshouldspecify:
whichvariablesyouareplanningtoanalyse;howtheyshouldbeanalysed;whatcross-tabulationsyouneedtoworkout;whichvariablesyouneedtocombinetoconstructyourmajorconceptsortodevelopindices(informulatingaresearchproblemconceptsarechangedtovariables–atthisstagechangethemback
toconcepts);whichvariablesaretobesubjectedtowhichstatisticalprocedures.
Toillustrate,letustaketheexamplefromthesurveyusedinthischapter.
FrequencydistributionsA frequency distribution groups respondents into the subcategories into which a variable can bedivided. Unless you are not planning to use answers to some of the questions, you should have afrequency distribution for all the variables. Each variable can be specified either separately orcollectivelyintheframeofanalysis.Toillustrate,theyareidentifiedhereseparatelybythenamesusedinthecodebook.Forexample,frameofanalysisshouldincludefrequencydistributionforthefollowingvariables:
AGEMS;TEDU(TEDU1,TEDU2,TEDU3–multipleresponses,tobecollectivelyanalysed);STUDY(STUDY1,STUDY2,STUDY3–multipleresponses,tobecollectivelyanalysed);DIFWK(DIFWK1,DIFWK2,DIFWK3,DIFWK4,DIFWK5–multipleresponses,tobecollectivelyanalysed);JOBSTA,JOBSTB;JOBSATA,JOBSATB;MOTIVA,MOTIVB.etc.
Cross-tabulationsCross-tabulations analyse two variables, usually independent and dependent or attribute anddependent,todetermineifthereisarelationshipbetweenthem.Thesubcategoriesofboththevariablesarecross-tabulatedtoascertain ifarelationshipexistsbetweenthem.Usually, theabsolutenumberofrespondents,andtherowandcolumnpercentages,giveyouareasonablygoodideaastothepossibleassociation.Inthestudywecitedasanexampleinthischapter,oneofthemainvariablestobeexplainedisthe
levelofsatisfactionwiththe‘before’and‘after’jobsafterredeployment.Wedevelopedtwoindicesofsatisfaction:
1. satisfactionwiththejobbeforeredeployment(SATINDB);2. satisfactionwiththejobafterredeployment(SATINDA);
Differencesinthelevelofsatisfactioncanbeaffectedbyanumberofpersonalattributessuchastheage,education,trainingandmaritalstatusoftherespondents.Cross-tabulationshelptoidentifywhichattributesaffectthelevelsofsatisfaction.Theoretically,itispossibletocorrelateanyvariables,butitisadvisabletobeselectiveoranenormousnumberoftableswillresult.Normallyonlythosevariablesthatyouthinkhaveaneffectonthedependentvariableshouldbecorrelated.Thefollowingcross-tabulationsareanexampleofthebasisofaframeofanalysis.Youcanspecifyasmanyvariablesasyouwant.
SATINDAandSATINDB
AGE;MS;TEDU;STUDY;DIFWK.
These determine whether job satisfaction before and after redeployment is affected by age, maritalstatus,education,andsoon.
SATINDAbySATINDBThisascertainswhetherthereisarelationshipbetweenjobsatisfactionbeforeandafterredeployment.
ReconstructingthemainconceptsThere may be places in a research instrument where you look for answers through a number ofquestionsaboutdifferentaspectsofthesameissue,forexamplethelevelofsatisfactionwithjobsbeforeandafterredeployment(SATINDBandSATINDA).Inthequestionnairetherewere10aspectsofajobaboutwhichrespondentswereaskedtoidentifytheirlevelofsatisfactionbeforeandafterredeployment.Thelevel of satisfaction may vary from aspect to aspect. Though it is important to know respondents’reactionstoeachaspect,itisequallyimportanttogaugeanoverallindexoftheirsatisfaction.Youmusttherefore ascertain, before you actually analyse data, how you will combine responses to differentquestions.In this example the respondents indicated their level of satisfaction by selecting one of the five
responsecategories.Asatisfactionindexwasdevelopedbyassigninganumericalvalue–thegreaterthemagnitude of the response category, the higher the numerical score – to the response given by arespondent. The numerical value corresponding to the category ticked was added to determine thesatisfaction index. The satisfaction index score for a respondent varies between 10 and 50. Theinterpretationofthescoreisdependentuponthewaythenumericalvaluesareassigned.Inthisexamplethehigherthescore,thehigherthelevelofsatisfaction.
StatisticalproceduresInthissectionyoushouldlistthestatisticalproceduresthatyouwanttosubjectyourdatato.Youshouldidentifytheproceduresfollowedbythelistofvariablesthatwillbesubjectedtothoseprocedures.Forexample,Regressionanalysis:
SATINDAandSATINDB
Multipleregressionanalysis:
Analysisofvariance(ANOVA):Similarly,itmaybenecessarytothinkaboutandspecifythedifferentvariablestobesubjectedtothe
variousstatisticalprocedures.Thereareanumberofuser-friendlyprogramssuchasSPSSxandSASthatyoucaneasilylearn.
Analysingquantitativedatamanually
Codeddatacanbeanalysedmanuallyorwiththehelpofacomputer.Ifthenumberofrespondentsisreasonablysmall,therearenotmanyvariablestoanalyse,andyouareneitherfamiliarwitharelevantcomputerprogramnorwishtolearnone,youcanmanuallyanalysethedata.However,manualanalysisisusefulonlyforcalculatingfrequenciesandforsimplecross-tabulations.Ifyouhavenotentereddataintoacomputerbutwanttocarryoutstatisticaltests,theywillhavetobecalculatedmanually,whichmaybecomeextremelydifficultandtimeconsuming.However,theuseofstatisticsdependsuponyourexpertiseanddesire/needtocommunicatethefindingsinacertainway.Be aware that manual analysis is extremely time consuming. The easiest way to analyse data
manuallyistocodeitdirectlyontolargegraphpaperincolumnsinthesamewayasyouwouldenteritinto a computer. On the graph paper you do not need to worry about the column number. Detailedheadingscanbeusedorquestionnumberscanbewrittenoneachcolumntocodeinformationaboutthequestion(Figure15.8).Toanalysedatamanually(frequencydistributions),countvariouscodesinacolumnandthendecode
them.Forexample,agefromFigure15.8,5=1,6=2.Thisshowsthatoutofthethreerespondents,onewasbetween40and44yearsofageandtheothertwowerebetween45and49.Similarly,responsesforeach variable can be analysed. For cross-tabulations two columns must be read simultaneously toanalyseresponsesinrelationtoeachother.Ifyouwanttoanalysedatausingacomputer,youshouldbefamiliarwiththeappropriateprogram.
Youshouldknowhowtocreateadatafile,howtousetheproceduresinvolved,whatstatisticalteststoapplyandhowtointerpretthem.Obviouslyinthisareaknowledgeofcomputersandstatisticsplaysanimportantrole.
FIGURE15.8Manualanalysisusinggraphpaper
Parttwo:Dataprocessinginqualitativestudies
Howyouprocessandanalysedatainaqualitativestudydependsuponhowyouplantocommunicatethe findings.Broadly, thereare threeways inwhichyoucanwrite aboutyour findings inqualitativeresearch: (1)developinganarrative todescribeasituation,episode,eventor instance; (2) identifyingthemain themes that emerge from your field notes or transcription of your in-depth interviews andwriting about them, quoting extensively in verbatim format; and (3) in addition to (2) above, alsoquantifythemainthemesinordertoprovidetheirprevalenceandthussignificance.Editing,asunderstoodforquantitativestudies,isinappropriateforqualitativeresearch.However,itis
possiblethatyoumaybeabletogothroughyournotestoidentifyifsomethingdoesnotmakesense.Insuchanevent,youmaybeabletorecallthecontextandcorrectthecontents,butbecarefulindoingsoas inability to recallpreciselymay introduce inaccuracies (recallerror) inyourdescription.Anotherway of ensuring whether you are truly reflecting the situation is to transcribe the interviews orobservationalnotesandsharethemwiththerespondentsorresearchparticipantsforconfirmationandapproval.Validationoftheinformationbyarespondentisanimportantaspectofensuringtheaccuracyofdatacollectedthroughunstructuredinterviews.Forwritinginanarrativeformatthereisnoanalysisperse,however,youneedtothinkthroughthe
sequenceinwhichyouneedorwanttonarrate.Fortheothertwowaysofwritingaboutthefindings,youneed to go through content analysis, asmentioned earlier.Content analysismeans analysing thecontentsofinterviewsorobservationalfieldnotesinordertoidentifythemainthemesthatemergefromtheresponsesgivenbyyourrespondentsortheobservationnotesmadebyyou.Thisprocessinvolvesanumberofsteps:
Step1
Identifythemainthemes.Youneedtogocarefullythroughdescriptiveresponsesgivenbyyourrespondentstoeachquestioninordertounderstandthemeaningtheycommunicate.Fromtheseresponsesyoudevelopbroadthemesthatreflectthesemeanings.Youwillnoticethatpeopleusedifferentwordsandlanguagetoexpressthemselves.Itisimportantforyoutoselectthewordingofyourthemesinawaythataccuratelyrepresentsthemeaningoftheresponsescategorisedunderatheme.Thesethemesbecomethebasisforanalysingthetextofunstructuredinterviews.Similarly,youneedtogothroughyourfieldnotestoidentifythemainthemes.
Step2
Assigncodestothemainthemes.Whetherornotyouassignacodetoamainthemeisdependentuponwhetherornotyouwanttocountthenumberoftimesathemehasoccurredinaninterview.Ifyoudecidetocountthesethemesyoushould,atrandom,selectafewresponsestoanopen-endedquestionorfromyourobservationalordiscussionnotesandidentifythemainthemes.Youcontinuetoidentifythesethemesfromthesamequestiontillyouhavereachedsaturationpoint.Writethesethemesandassignacodetoeachofthem,usingnumbersorkeywords,otherwisejustidentifythemainthemes.
Step3
Classifyresponsesunderthemainthemes.Havingidentifiedthethemes,thenextstepistogothroughthetranscriptsofallyourinterviewsoryournotesandclassifytheresponsesorcontentsofthenotesunderthedifferentthemes.YoucanalsouseacomputerprogramsuchasEthnograph,NUD*ISTN6,NVivo,XSightforundertakingthisthematicanalysis.Youwillbenefitbylearningoneoftheseprogramsifyourdataissuitableforsuchanalysis.
Step4
Integratethemesandresponsesintothetextofyourreport.Havingidentifiedresponsesthatfallwithindifferentthemes,thenextstepistointegratethemintothetextofyourreport.Howyouintegratethemintoyourreportismainlyyourchoice.Somepeople,whilediscussingthemainthemesthatemergedfromtheirstudy,useverbatimresponsestokeepthe‘feel’ofthe
responses.Thereareotherswhocounthowfrequentlyathemehasoccurred,andthenprovideasampleoftheresponses.Itentirelydependsuponthewayyouwanttocommunicatethefindingstoyourreaders.
Contentanalysisinqualitativeresearch–anexample
The above four-step processwas applied to a study recently carried out by the author to develop anoperational service model, based upon the principle of family engagement. The information waspredominantly gathered through in-depth and focus group discussionswith clients, service providersandservicemanagers.Afterinformaltalkswithanumberofstakeholders,alistofpossibleissueswasdevelopedtoformthebasisofdiscussionsinthesein-depthinterviewsandgroupdiscussions.Thelistwasmerelyaguidingframeworkandwasopentoinclusionofanynewissuethatemergedduringthediscussions.Outoftheseveralissuesthatwereidentifiedtoexaminevariousaspectsofthemodel,herethe author has taken only one to show the process of identifying themes that emerged during thediscussions.Notethatthesethemeshavenotbeenquantified.Theyaresubstantiatedasverbatim,whichis one of themain differences between qualitative and quantitative research. The following exampleshowsperceivedstrengthsof theFamilyEngagementModel (FEM)as identifiedby the stakeholdersduringin-depthinterviewsandfocusgroups.InformationprovidedinFigure15.8providesanexampleoftheoutcomeofthisprocess.
Example:Developingthemesthroughcontentanalysis
Perceivedstrengthsofthemodel
Theframeworkdevelopedfortheperceivedstrengthsofthemodelisbasedupontheanalysisoftheinformationgathered,whichsuggestedthatthevariousthemesthatemergedduringthedatacollectionstagereflectingstrengthsofthemodelcanbeclassifiedunderfourperspectives.Thefollowingdiagramshowstheframeworkthatemergedfromtheanalysis.
Differentperspectivesclassifyingperceivedstrengthsofthemodel.
PerceivedstrengthsfromtheperspectiveofthefamilyThissectiondetailstheperceivedstrengthsofthemodelfromtheperspectiveofthefamily.Keepinmindthatthesequentialorderoftheperceivedstrengthsisrandomanddoesnotreflectanyorderorpreference.Also,thenamingofthesethemesisthatoftheauthor,whichtothebestofhisknowledgecapturedthe‘meanings’oftheintentionoftheresearchparticipants.
Empowermentoffamilies
Almosteveryoneexpressedtheopinionthatoneofthemainstrengthsofthemodelisthatitempowersfamiliesandclientstodealwiththeirownproblems.Themodelprovidesanopportunitytofamiliestoexpresstheirfeelingsaboutissuesofconcerntothem
and,tosomeextent,totakecontroloftheirsituationsthemselves.Itseemsthatin‘preparingaplanforachildunderthismodel,thefamilyofthechildwillplayanextremelyimportantroleindecidingaboutthefutureofthechild,whichisthegreateststrengthofthemodel’.Oneoftherespondentsexpressedhis/heropinionasfollows:
Oh, the Family Engagementmodel actually gives the power back to the family but with the bottom line in place, like theDepartment’sbottom lines, theyhave tomeet them.Oh… theoldmodelwouldhavebeenblackandwhite; kids remain inMum’s care, he (the father) would have supervised contacts with kids and it all would have been set up … the FamilyEngagementmodelwasaboutpullingtheminthewholefamily thencomingupwith thesolutionsas longas theyreachtheDepartment’sbottomline.Theyactuallyhavetocomeupandnominatewhattheywerewillingtodo…He(father)returnedhome,whichwasmuchbetter…Iftheyhaverelapsewebringthembackinandwetalkaboutit,getthembackontrack,makesuretheywereengagingwiththeservices…Intheoldmethod,kidsjustwouldhavebeenremovedandkidswouldhavegoneintotheDepartment’scare…Itismoreempoweringtothefamily,anditismucheasiertoworkwiththefamilyatthatlevelthanyouarestandingoverandtellingthemthatyouhavetodothisandthis,andholdingitagainstthemthatifyoudonotdo,well,thekidsareout.Itismuch,muchbetterforthefamilies.You’vegotmoreopportunitytoworkwiththefamilyatthatlevel,ratherthanbeingontheoutsidedictating.
Anotherparticipantsaid:
Ithinkthismodelempowersthefamilyalotmore…youarehavingmeetingsallthetime.Yougivethemthebottomline,andtheydeveloptheirownstrategy…Ithinkitempowersthefamilywhentheycomeback…becausetheyaredevelopingtheirplan,theyareusingtheirownnetworkandresources…Ithinkitisempowering.
Yet,accordingtoanotherrespondent:‘Itallowsthemtofeelthattheycanmakesomedecisions…They are able to work with the Department and that their voices or views are as valid as theDepartment’s.’
BuildingofcapacityoffamiliesAnother advantage that cameout of the discussions is that the process adopted as a part of theFEMmakesclientsawareofcommunityresourceswhich,inturn,helpthemtobuildtheircapacitytodealwithasituationeffectivelyandindependently.Asoneparticipantpointedout:
Theyknowthat,ok,ifsomethinggoeswronginthisaspectoftheirlife,theyknowtheycangothereforsupport,theydonotneedtobecallingus…theymayhaveresolvedtheirownissues…thatisreallyempowering.
Anotherparticipantsaid:‘Underthismodel,afamilyhastakenamuchstrongerroleinbringingaboutchangeascomparedtothecaseconferenceapproach.’
AcknowledgementofpositivesinfamiliesOneofthestrengthsoftheapproachisthatitacknowledgesthestrengthsoffamilies.Themodelisprimarilybasedupondesigning interventionsbasedon the strengths andpositivesof a situationrather than on the negatives. ‘In the oldmodel the strengthswere not acknowledged to a largedegreeandcertainlynotoftheparents.’Duringoneofthefocusgroups,aparticipantexpressedtheviewsthat:
TheFamilyEngagementmodelstartswiththestrengthsofthefamily,sobringingthefamilyinatanearlierstage,andtryingtogetthemtohelpmakedecisionsaboutwhatisgoingtohappentochildrenwhoareincrisis…so…itisinvolvingmorepeople,theirextendedfamily,andgettingthemtocomeupwithaplan.
Inanotherfocusgroup,arespondentsaid:
Itisonlybecauseyouactuallydoworkwiththatstrength-basedapproachandyouacknowledgeit.Itisahugepartofwhathappens.Youcanactuallysaytosomebodythatyouaredoingsowell,itisgreattoseethechangeinyou,andeventhoughyoupersonallyhavenothingtodowiththosechanges,yousay,welldone…itissogoodtoseeyoulookingsowell…Yougettoapointinaprocesswhereyouarenolongerseenasanoutsider,youarenolongerseenasaprescriptiveorganisation,butyouareseenasasupportiveorganisationwhichisactuallyassistingthatpersonintheprocess…ThisDepartmenthasnotbeengoodinacknowledgingchange,wehavenotbeen.
CollaborativedecisionmakingandsolutionsAnotherstrengthpointedoutbymanyisthatsolutionspertainingtoachildarenowdevelopedin
close collaboration with the family, extended family and other appropriate stakeholders, whichmakes them (decisions and solutions)more acceptable andworkable. In oneof the focusgroupinterviewsaparticipantexpressedthisstrengthofthemodelasfollows:
Theycomeupwithanswers,theygotit.Youareupfrontwiththefamily.Itgivesthefamilyaveryclearideawhatexactlyisexpectedof them… this iswhat theyhavedone,what are the concernsof theDepartment…Sobyhaving family supportmeetings,youaretellingthefamilythisiswhatwefeelishappeningwiththechildandthesearethethingsthatcannothappentoyourchildren,whatweintendtodoaboutitinthefuturetobeabletohavethembackortoimprovetheirenvironment…itisstraightinfrontofthem,notbehindtheirback…PreviouslyIknowofacasewhereafamilywasnotinvolvedinanyofthediscussionsandtheydidnotunderstandwhytheirchildrenwereremovedfromthem.
Manyparticipantsfeltverypositiveaboutthiscollaborativeapproach.Theyfeltthat,‘havingafamilysupportmeetingclearlytellsthemwhathashappened,whataretheintentionsoftheDepartment,andhowtheDepartmentisgoingtoworkwiththem’.Anotherparticipantinoneofthefocusgroupdiscussionsadded:
Therearedifferencesbetweenhowthemeetingsareheldbut,Iguess,oh,sometimestogetthefamilytodevelop,andremindthemof the bottom line, rather thanus saying, ‘This iswhat Iwant to happen.’ Imean, obviously in the discussionof thegeneralsituation,youmakethingsclear,butyouletthefamilytaketheresponsibilitytodeveloptheirownplan.
Arespondent inan in-depth interviewexpressed theopinion that,now, ‘Weare identifying themembersof theextendedfamily.Onceupon-a-timewejusthadparents;nowyouhavetogoaroundandsearchandgetthemalltogethertomakeadecision’.Yet another respondent, talking about the strengths of themodel, said, ‘Nowwe approach verydifferently.’Accordingtohim/her:
Weinformyou,weadviseyou,thatthechildrenareatrisk…whateverwiththechildren,wewanttositdownandworkwithyou.Throughoutthisprocesswewanttoworkwithyou,andalso,planshavebeensetup.Wewantyoutobeapartofthat.
It appears that, under this model, decisions are made not by a single individual, but by all those involved. According to onerespondent,‘Youaresharingresponsibilitywithotheragenciesandfamilymembers;itisnotonlyyourdecision,itisthedecisionofeverybody.’
KeepsfamiliesintactSome respondents also felt, that in certaincases, ‘Thechildrenmaynot evenbe taken from thefamily so quickly.’ It seems there is a greater attempt to keep the children in the family. Oneparticipant said: ‘I amactuallyworkingwithquite a fewkidswhere theyare trying tokeep thefamily together.’ Another participant added that, ‘The apprehension rate has come downsubstantially.’
PerceivedstrengthsfromtheperspectiveofthechildAgreaterfocusonchildrenManyfamilyworkersaswellasteamleadersfeltthatthewholeapproachisalotmorefocusedonchildren.Theapproachischildcentredand,ateverystep,concernsforchildrenformthecoreofanintervention: ‘It is a lot more child-focused as well. Rather than focusing on the parents, it isfocusingonhowwearegoingtomakethischildsafe,andhowwearegoingtoachievethat.’
ReturnschildrentotheirparentsquickerSomerespondentsfeltthatthenewapproachhelpedchildrentogetbacktotheirparentsquicker.Inoneofthefocusgroupdiscussions,oneparticipantsaid:
Ithinkitgetsthechildrenbacktotheirparentsquickerbecauseatthemeetingitidentifiesstrongpeopleinthefamilythatcansupporttheparentstokeeptheirkids.SowhatIfoundintheofficeisthatsomeofthekidsgetbacktotheirparentsquickerthanthroughtheCaseConference.TheCaseConferenceiseveryyear…whatthefamilieshavetojumpthroughbytheendofthetwelvemonthsatthenextmeeting…hereitisnoneofthat.Itisastrongperson,howyouaregoingtosupporttheMumtokeepgettingthekidsback…whatyoudoneed…sometimesthekidsgoback,justlikethat.
PreventionofremovalofchildrenSome respondents felt that the model actually prevented kids from being removed from theirfamilies.According to one participant, ‘that is the big advantage of thismodel; to prevent kidsfrombeingremoved’.
Perceivedstrengthsfromtheperspectiveofserviceproviders
GreaterjobsatisfactionAlmosteveryserviceprovidersaidthattheirworkaftertheintroductionofthemodelhadbecome‘muchmoresatisfying’because‘itisenablingworkersandclients’.
EasierfortheworkerstoworkunderthemodelMany respondents felt that ‘initially it ismorework for aworkerbut, in the long run, it is lessworkbecauseofthesharedresponsibility’.Accordingtooneparticipant:
Asa caseworker, I remember thedayswhen I had to reallywork sohard tomeet somanypeopleanddo somany thingsindividuallyandalltheresponsibilitywasonmyshoulders…butnowthereisasharedresponsibility…youhavetodothegroundworkbutwhenitisdonethelongtermengagementiseasierbecausetherearemorepeopleinvolvedandtheyinpartmakethedecisions.
DeclineinhostilitytowardstheDepartmentAnother advantage that some respondents pointedoutwas a decline in hostility among familiestowardstheDepartment.Itwaspointedoutthatthoughitdependeduponthecircumstances,therewas a feeling that, on thewhole, hostility among clients towards theDepartment had declined.They also felt that though, in thebeginning, theremight havebeenhostility,workingunder thenewmodel, inmost situations,made that hostility disappear. In some situations, an increase inhostility was possibly attributable to a situation such as the apprehension of a child. Mostrespondents were of the opinion that, as compared to the Case Management model, they hadexperiencedfarlesshostilitytowardsthemwhileworkingunderthenewapproach.
IncreasedtrustinthestaffbyclientsAnother advantage some workers saw in the new approach was that they felt that clients hadstartedtrustingthemalotmore.Aparticipantinoneofthefocusgroupsdescribedhis/herfeelingsasfollows:
Theycallusnow,Idonothavetogoandlookforthem.Theyarecallingmenowandaskingwhatisthis?…whichmeanstheyaretakinganinterest.Theyarenotsittingbackandsaying,ohwell,theyaregoingtotellusornottellus.
BetterrapportwithfamiliesBecause of the increased number of contacts, it appears that staff were able to build morecongenialandtrustingrelationshipswithfamilies.Inafocusgroup,oneoftherespondentssaid,‘Ithinktherelationshipismorerespectfulandtrustful’.Anotherrespondentsaid:
Familyrelationshipsarealittlebitbetter,andafamilyalsounderstandsthataDCDworkerisnotsomeonewhogoestohomesandremovestheirchildren…howhorriblepeopleweare,but,byinteractingwiththemtheyactuallyunderstandthatwearepeopleatwork,andthatwearenotgoingtodothesethings,thatistheoldwayofdoing.Wearenotgoingtoremoveachildwithoutsayinganything.Wehavecommunicationwiththem.
Developsbetterunderstandingbyworkersofthefamilydynamics‘One of the strengths that I have seen is that it allows the social worker to feel the family
dynamics, to thinkabout thedynamicsand it allows the families toparticipate inwhatever theywantto’,saidaparticipantofafocusgroup.
FeweraggressiveclientsAnotherobviousdifferencebetween the twomodels, according to some respondents,was feweraggressiveclients.Onerespondentsaid,
Therearefarlessaggressiveclientshereascomparedtootherplaces.Ithinkitispartlybecauseoftheapproach.Hereyouveryrarelyseepeoplewhogetagitated,itismuchmorecontrolled,anditisacalmeratmosphere.
DevelopsasenseofownershipofacaseOneoftheadvantagesoftheFEMaspointedoutbyvariousrespondentswasthatunderthemodel,‘you feel the case belongs to you; you “own” a child’. Because of this, according to someparticipants,therewasagreateraffiliationbetweenthefamilyworkerandthechild.
GreatercommunityinteractionAnother advantagepointedout by familyworkerswas that themodel resulted in their having agreater interaction with community agencies and, consequently, had more knowledge of theircommunityandtheservicesavailableinit.ThiswasprimarilybecauseofrestructuringundertheFamilyEngagementmodelwherebyfamilyworkerswereallocatedparticulargeographicalareas,called,‘patch’:‘Youalsodevelopreallyniceworkingrelationswiththosepeople.Youareworkingtogether collaboratively towards the goals, and I think, that is really a great benefit’, said oneparticipant.Therewas‘alotmorelinkingwithotheragencies’underthemodel.Notonlywastheinteractionbetweencommunityagenciesincreasedbut,itappears,clientshadalsostartedmakingmoreuseofcommunityagencies.
GreaterknowledgeaboutcommunitymembersAnotherbenefitofworkingunder theFEMandwithina ‘patch’was that familyworkersgot toknowalotmorecommunitymembers.Accordingtoonerespondent,‘Therelationshipwithpeopleinyourcommunityismuchstrongerandwidespread.’Anotherrespondentsaid:
Afteracertaintimeyougettoknowwholivesonwhatstreet,familylinksbetweenpeople,especially,whenyouareworkingwithAboriginalfamilies.Familylinksaresoimportant,andknowingwhoisdealingwithwhom…knowingwhoisinthearea,whatresourcesyouhave,makesyourjobalotmoreeffective.
GreatercontroloverpersonalvaluesbyworkersAnother advantage identifiedby some respondentswas that,with thenewmodel, ‘caseworkersownvaluesandmoralscannotbeimposed’.
Perceivedstrengthsfromtheperspectiveoftheservicedeliverymanner
Anopen,honestandtransparentprocessThewholeprocessisopentoallstakeholders.‘Allthecardsareonthetable’,saidoneparticipant,andanotherexpressedtheopinionthat:‘Thecaseworkermaybehonest,but,Iguess,theprocess,howitwasdone,wasnot.’Onerespondentsaidthatoneofthegoodthingsaboutthismodelwasthat‘everyoneknowswhatisgoingon’.Accordingtoarespondent:
Anothergoodthingaboutthesefamilymeetingsisthatthereistheparents,thereisthefamily.TheparentsmighthavebeentellingusonethingorapartofthestoryandJaime,anotherpartofthestory,nottellingUncleJimmy…soitisgoodinawaythateveryoneknowswhatisgoingon.Thewholeinformationisthereforeveryonethatisthere.Sotheycannotpushittous
and push it to the family. That is another good thing: they all get the same information, and we get and give the sameinformationtothem…anditisamazingtheplantheywanttocomeupwith…itisbasedupontheinformationgiventothem.
Anotherrespondentinafocusgroupsaid:
Andyouareactuallyfightingtheparentsabouttheguardianshipofthechild:attheendofthedaythatiswhatyouaredoing,and, I think, just to have the transparent working relationship within the Family Engagement model actually makes thatprocessaloteasierbecauseeverythingisoutintheopenandwhenitcomesupinthecourt,theyarenotgoingtobesurprised.
GreaterinformalityinmeetingsFamily meetings under the FEM are far less formal: ‘The family members and others areencouragedtosaywhatevertheywantto.Theycaninterruptandstopthechairpersonanytime,iftheydisagree.Theycanevencomebacklater.’‘Whatisimportantisthattheminutesarewrittenup,andthefamilygetsacopyoftheminutessothattheycangobackhomeandreadtheminutes.Theycancomebacktous.’
MorefrequentreviewofcasesManypeoplefelt that themodelprovidedanopportunity toreviewcasesmorefrequentlywhichhelpedthemtoachievegoalsmorequicklyand,ifaninterventionwasnotworking,ithelpedthemto change the intervention. As one participant pointed out, ‘Changes in the plan to reflect thechangesinthefamilydynamicsareundertakenfrequently.’Hence,underthemodel,‘Theplanfora child is continually being reviewed.’ There seemed to be a lot more flexibility in terms ofchangingaplanunderthemodel.
Increasedhonesty,transparencyandaccountabilitySome respondents also felt that, because of the transparency and accountability of the process,simply working within the parameters of the model had helped to keep workers honest andaccountable.Accordingtooneparticipant,‘Fromapracticeviewpoint,itallowsthesocialworkertobehonest,accountableandtobetransparent.’
AfairerapproachMany respondents felt that the FEMwas fairer, as itwas open, participatory and empowered afamily.
GoalssetforclientsaremoreattainableandworkableAccording to one of the participants of a focus group: ‘I think the plan of FamilyEngagementmeetingsismoreattainableandmoreworkable…whattheyareactuallycapableofdoing.Wearenotsettingwhattheyarenotgoingtoachieve,sotheyarenotgoingtofail.’Inaddition,itseems,becausefamilieswereinvolvedindevelopingaplan,theyhadafeelingofownership,andhencetheyattainedthetaskssetoutinit.Anotherparticipantwasoftheopinionthat:‘Ifyouareapartofthesolution,thenyouactuallyhaveaninvestmentinmakingthechange.’
EqualityinrelationtoexpressionofopinionSomerespondentsfeltthatthemodelprovidedfreedomofexpressiontoparties.Allinvolvedwerefree to express their opinions, and they were encouraged to share their views. As long as thebottomlinewasmet,theiropinionsweretakenintoconsiderationindevelopingastrategy.
AlesschaoticprocessAsoneparticipantobserved:‘Itisfarlesschaotic,justtheperceptionofwhatwasgoingon.They[referringtoworkersintheCMM]feltabitchaoticbecauseworkwascominginallthetimeand
theywereholdingontocases.Hereitismoreorganised’,oneparticipantobserved.Withtheoldstructure,‘caseworkerswereverystressed;theywerenotoperatingparticularlywell’.
AlessstressfulapproachManyparticipantsfeltthatthenewapproachwaslessstressfulbecauseofitsmanybenefits.Itwasless stressful for them, and for families, as well as children. In one of the focus groups, aparticipantexpressedhis/herfeelinginthefollowingwords:
YoudonotfeelthatIhatetogotothishome…howaretheygoingtoreact,whataretheygoingtosaytome,orhowshouldIleaveor how should I protectmyself?Youdonot have to have those stresses now; it is a calmer situation, it is a happiersituationand that is good for the kids, not only forus, but for the kids… it is actually the kidswhoalsobenefit from theapproach.
FewerconflictswithfamiliesManyrespondentsfeltthatongoingconflictswithfamilieswerefarfeweraftertheintroductionofthenewmodel.
EqualityregardingchoiceofafacilitatorformeetingsAnother strengthwas that someparticipants thought that under thenewmodel facilitationworkwasnotonlyconfinedtocasemanagers.Underthemodel,anyonecouldbecomeafacilitator.
IncreasedreflectiononpracticeSomepeoplealso felt that themodelprovidedanopportunity to reflectonpractice thushelpingthemtoimproveit.
TotalresponsibilityforcasesSome also pointed out that workers have the total responsibility for cases which seemed to bemuchbetterfromanumberofviewpoints.Aspointedoutbyoneperson,‘Underthemodel,afieldworkerisresponsibleforthetotalintervention,fromAtoZ.Youdoeverythinginapatch.’
Compliancewithgovernment’schildplacementpolicyOne of the participants pointed out that the model actually complied with the government’slegislative obligation to placeAboriginal andTorres Strait Islander childrenwith their families.Accordingtothisparticipant:
Themodelactuallymeets,forAboriginalandTorresStraitIslanderschildplacementprinciplewhichisnowenshrinedinourlegislation, where it actually states that children will be placed with family, extended family, immediate community andextendedcommunityandanon-Aboriginalpersonisalastresort…Sothismodelactuallymeetsthat.
Theroleofstatisticsinresearch
Theroleofstatistics inresearchissometimesexaggerated.Statisticshavearoleonlywhenyouhavecollectedtherequiredinformation,adheringtotherequirementsofeachoperationalstepoftheresearchprocess.Oncedataiscollectedyouencountertwoquestions:
1. HowdoIorganisethisdatatounderstandit?2. Whatdoesthedatamean?
In away, the answer to the first question forms the basis for the second. Statistics can play a veryimportant role in answering your research questions in such amanner that you are able to quantify,measure, place a level of confidence on the findings, make an assessment of the contribution eachvariable hasmade in bringing out change,measure the association and relationship between variousvariables,andhelppredictwhatislikelytohappeninthelightofcurrenttrends.From individual responses, particularly if there are many, it becomes extremely difficult to
understand the patterns in the data, so it is important for the data to be summarised. Some simplestatisticalmeasuressuchaspercentages,means,standarddeviationsandcoefficientsofcorrelationcanreduce the volumeof data,making it easier to understand. In computing summarymeasures, certaininformation is lost and therefore misinterpretation is possible. Hence, caution is required wheninterpretingdata.Statisticsplayavitalroleinunderstandingtherelationshipbetweenvariables,particularlywhenthere
aremorethantwo.Withexperience,it iseasyto‘read’therelationshipbetweentwovariablesfromatable,butnottoquantifythisrelationship.Statisticshelpyoutoascertainthestrengthofarelationship.Theyconfirmorcontradictwhatyoureadfromapieceofinformation,andprovideanindicationofthestrengthoftherelationshipandthelevelofconfidencethatcanbeplacedinfindings.Whentherearemorethantwovariables,statisticsarealsohelpfulinunderstandingtheinterdependencebetweenthemandtheircontributiontoaphenomenonorevent.Indirectly,knowledgeofstatisticshelpsyouateachstepoftheresearchprocess.Knowledgeofthe
problemsassociatedwithdataanalysis,thetypesofstatisticaltestthatcanbeappliedtocertaintypesofvariable,and thecalculationofsummarystatistics inrelation to themeasurementscaleusedplaysanimportantroleinaresearchendeavour.However,youcanalsocarryoutaperfectlyvalidstudywithoutusinganystatisticalprocedures.Thisdependsupontheobjectivesofyourstudy.
Summary
Inthischapteryouhavelearntaboutprocessingdata.Irrespectiveofthemethodofdatacollection,qualitativeorquantitative, theinformationiscalled‘rawdata’orsimply‘data’.Theprocessingofdata includesalloperationsundertakenfromwhenasetofdata iscollecteduntil it is ready tobeanalysedeithermanuallyorbyacomputer.Dataprocessinginquantitativestudiesstartswithdataediting,whichisbasically‘cleaning’yourdata.Thisisfollowedbythecodingofdata,whichentailsdevelopingacodebook,pre-testingit,codingperseandverifyingthecodeddata.Inthischapterwehave provided a prototype for developing a code book, detailing descriptions of how to developcodes for open-ended and closed questions, and a step-by-step guide to coding data, taking anexamplefromasurvey.Thechapteralso includesdetailed informationaboutcontentanalysisandhow to treat data for narrative and thematic styles of writing, and an extended example from aqualitativestudyisprovided.Though thedevelopmentofa frameofanalysis continuesuntilyouhave finished the report, it
helps immensely indataanalysis todevelopthisbeforeyoubeginanalysingdata. In theframeofanalysisthetypeofanalysistobeundertaken(e.g.frequencydistribution,cross-tabulation,contentanalysis),andthestatisticalprocedurestobeapplied,shouldbespecified.Computers primarily help by saving labour associated with analysing data manually. Their
application in handling complicated statistical and mathematical procedures, word processing,displayingandgraphicpresentationoftheanalyseddatasavestimeandincreasespeed.Statisticsare
desirablebutnotessentialforastudy.Theextentoftheirapplicationdependsuponthepurposeofthestudy.Statisticsprimarilyhelpyoutomakesenseofdata,‘read’thedata,explorerelationshipsand the interdependencebetweenvariables, ascertain themagnitudeof anexisting relationshiporinterdependenceandplaceconfidenceinyourfindings.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Whatprocedurescanyousetinplacetoensuretheaccuracyoftheinformationobtainedinbothquantitativeandqualitativestudies?Thinkingofexamplesfromyourownareaofstudy,considertheadvantagesanddisadvantagesofhavingusedopen-endedorclosedquestionswhenyoucometoprocessyourdata.Assesstheroleofstatisticsforastudyinyourareaofinterest.
CHAPTER16DisplayingData
Inthischapteryouwilllearnabout:
MethodsofcommunicatinganddisplayinganalyseddatainquantitativeandqualitativeresearchHowtopresentyourdataintablesDifferenttypesofgraphsandhowtousethemtorepresentyourdata
Keywords:areachart,bardiagram,bivariate,cumulativefrequencypolygon,datadisplay,frequencygraph,linediagram,piechart,polygon,polyvariate,scattergram,table,univariate.
Methodsofcommunicatinganddisplayinganalyseddata
Having analysed the data that you collected through either quantitative or qualitativemethod(s), thenexttaskistopresentyourfindingstoyourreaders.Themainpurposeofusingdatadisplaytechniquesis to make the findings easy and clear to understand, and to provide extensive and comprehensiveinformationinasuccinctandeffectiveway.Therearemanywaysofpresentinginformation.Thechoiceofaparticularmethodshouldbedeterminedprimarilybyyour impressions/knowledgeofyour likelyreadership’sfamiliaritywiththetopicandwiththeresearchmethodologyandstatisticalprocedures.Ifyour readers are likely to be familiarwith ‘reading’ data, you can use complicatedmethods of datadisplay;ifnot,itiswisetokeeptosimpletechniques.Althoughtherearemanywaysofdisplayingdata,thischapter is limitedtothemorecommonlyusedones.Therearemanycomputerprogramsthatcanhelpyouwiththistask.Broadly,therearefourwaysofcommunicatinganddisplayingtheanalyseddata.Theseare:
1. text;2. tables;3. graphs;and4. statisticalmeasures.
Becauseofthenatureandpurposeofinvestigationinqualitativeresearch,textbecomesthedominantand usually the sole mode of communication. In quantitative studies the text is very commonlycombinedwithotherformsofdatadisplaymethods,theextentofwhichdependsuponyourfamiliaritywith them, thepurposeof the study andwhat you thinkwouldmake it easier for your readership tounderstandthecontentandsustaintheir interest init.Henceasaresearcherit isentirelyuptoyoutodecidethebestwayofcommunicatingyourfindingstoyourreaders.
Text
Text,byfar,isthemostcommonmethodofcommunicationinbothquantitativeandqualitativeresearchstudies and, perhaps, the only method in the latter. It is, therefore, essential that you know how tocommunicateeffectively,keepinginviewthelevelofunderstanding,interestinthetopicandneedforacademic and scientific rigour of those forwhomyou arewriting.Your style should be such that itstrikes a balance between academic and scientific rigour and the level that attracts and sustains theinterestofyourreaders.Ofcourse,itgoeswithoutsayingthatareasonablecommandofthelanguageandclarityofthoughtareimperativeforgoodcommunication.Your writing should be thematic: that is, written around various themes of your report; findings
should be integrated into the literature citing references using an acceptable systemof citation; yourwritingshouldfollowalogicalprogressionofthought;andthelayoutshouldbeattractiveandpleasingtotheeye.Language,intermsofclarityandflow,playsanimportantroleincommunication.AccordingtotheCommonwealthofAustraliaStyleManual(2002:49):
Thelanguageofwell-writtendocumentshelpstocommunicateinformationeffectively.Languageis also the means by which writers create the tone or register of a publication and establishrelationshipswith their readers.For these relationships tobeproductive, the language thewriterusesmust take full account of the diversity of knowledge, interests and sensitivitieswithin theaudience.
Tables
Structure
Other than text, tables are themost commonmethod of presenting analysed data.According toTheChicago Manual of Style (1993: 21), ‘Tables offer a useful means of presenting large amounts ofdetailed information in a small space.’ According to the Commonwealth of Australia Style Manual(2002:46),‘tablescanbeaboonforreaders.Theycandramaticallyclarifytext,providevisualrelief,and serve as quick point of reference.’ It is, therefore, essential for beginners to know about theirstructureandtypes.Figure16.1showsthestructureofatable.Atablehasfiveparts:
1. Title–Thisnormallyindicatesthetablenumberanddescribesthetypeofdatathetablecontains.Itisimportanttogiveeachtableitsownnumberasyouwillneedtorefertothetableswheninterpretinganddiscussingthedata.Thetablesshouldbenumberedsequentiallyastheyappearinthetext.Theprocedurefornumberingtablesisapersonalchoice.Ifyouarewritinganarticle,simplyidentifyingtablesbynumberissufficient.Inthecaseofadissertationorareport,oneway
toidentifyatableisbythechapternumberfollowedbythesequentialnumberofthetableinthechapter–theprocedureadoptedinthisbook.Themainadvantageofthisprocedureisthatifitbecomesnecessarytoaddordeleteatablewhenrevisingthereport,thetablenumbersforthatchapteronly,ratherthanforthewholereport,willneedtobechanged.Thedescriptionaccompanyingthetablenumbermustclearlyspecifythecontentsofthattable.Inthedescriptionidentifythevariablesaboutwhichinformationiscontainedinthetable,forexample‘Respondentsbyage’or‘Attitudestowardsuraniummining’.Ifatablecontainsinformationabouttwovariables,thedependentvariableshouldbeidentifiedfirstinthetitle,forexample‘Attitudestowardsuraniummining[dependentvariable]bygender[independentvariable]’.
2. Stub–Thesubcategoriesofavariable,listedalongthey-axis(theleft-handcolumnofthetable).AccordingtoTheMcGraw-HillStyleManual(Longyear1983:97),‘Thestub,usuallythefirstcolumnontheleft,liststheitemsaboutwhichinformationisprovidedinthehorizontalrowstotheright.’TheChicagoManualofStyle(1993:331)describesthestubas:‘averticallistingofcategoriesorindividualsaboutwhichinformationisgiveninthecolumnsofthetable’.
3. Columnheadings–Thesubcategoriesofavariable,listedalongthex-axis(thetopofthetable).Inunivariatetables(tablesdisplayinginformationaboutonevariable)thecolumnheadingisusuallythe‘numberofrespondents’and/orthe‘percentageofrespondents’(Tables16.1and16.2).Inbivariatetables(tablesdisplayinginformationabouttwovariables)itisthesubcategoriesofoneofthevariablesdisplayedinthecolumnheadings(Table16.3).
4. Body–Thecellshousingtheanalyseddata.5. Supplementarynotesorfootnotes–Therearefourtypesoffootnote:sourcenotes;othergeneral
notes;notesonspecificpartsofthetable;andnotesonthelevelofprobability(TheChicagoManualofStyle1993:333).Ifthedataistakenfromanothersource,youhaveanobligationtoacknowledgethis.Thesourceshouldbeidentifiedatthebottomofthetable,andlabelledbytheword‘Source:’asinFigure16.1.Similarly,otherexplanatorynotesshouldbeaddedatthebottomofatable.
Typesoftables
Depending upon the number of variables about which information is displayed, tables can becategorisedas:
FIGURE16.1Thestructureofatable
TABLE16.1Respondentsbyage(frequencytableforonepopulation–hypotheticaldata)
Age No.ofrespondents
<20years 2(2.0)20–24 12(12.0)25–29 22(22.0)30–34 14(14.0)35–39 17(17.0)40–44 10(10.0)45–49 11(11.0)50–54 9(9.0)55+00 3(3.0)Total 100(100.0)Note:Figuresinparenthesesarepercentages.
univariate(alsoknownasfrequencytables)–containinginformationaboutonevariable,forexampleTables16.1and16.2;bivariate(alsoknownascross-tabulations)–containinginformationabouttwovariables,forexampleTable16.3;andpolyvariateormultivariate–containinginformationaboutmorethantwovariables,forexampleTable16.4.
TABLE16.2Respondentsbyage(frequencytablecomparingtwopopulations–hypotheticaldata)
Note:Figuresinparenthesesarepercentages(*roundingerror).
TABLE16.3Respondentsbyattitudetowardsuraniumminingandage(cross-tabulation–hypotheticaldata)
*=columnpercentage;@=Rowpercentage.
Typesofpercentages
Theabilitiestointerpretdataaccuratelyandtocommunicatefindingseffectivelyareimportantskillsforaresearcher.Foraccurateandeffectiveinterpretationofdata,youmayneedtocalculatemeasuressuchaspercentages,cumulativepercentagesorratios.Itisalsosometimesimportanttoapplyotherstatisticalprocedurestodata.Theuseofpercentagesisacommonprocedureintheinterpretationofdata.Thereare three typesofpercentage:‘row’,‘column’and‘total’. It is important tounderstand therelevance,interpretationandsignificanceofeach.Letustakesomeexamples.
TABLE16.4Attitudetowardsuraniumminingbyageandgender(hypotheticaldata)
Tables16.1and16.2areunivariateorfrequencytables.Inanyunivariatetable,percentagescalculate
themagnitudeofeachsubcategoryofthevariableoutofaconstantnumber(100).Suchatableshows
whatwould have been the expected number of respondents in each subcategory had there been 100respondents.Percentagesinaunivariatetableplayamoreimportantrolewhentwoormoresamplesorpopulations are being compared (Table16.2).As the total number of respondents in each sample orpopulationgroupnormallyvaries,percentagesenableyoutostandardise themagainstafixednumber(100).Thisstandardisationagainst100enablesyoutocompare themagnitudeof the twopopulationswithinthedifferentsubcategoriesofavariable.In a cross-tabulation such as in Table 16.3, the subcategories of both variables are examined in
relationtoeachother.Tomakethistablelesscongested,wehavecollapsedtheagecategoriesshowninTable16.1.Forsuchtablesyoucancalculatethreedifferenttypesofpercentage,row,columnandtotal,asfollows:
Rowpercentage–Calculatedfromthetotalofallthesubcategoriesofonevariablethataredisplayedalongarowindifferentcolumns,inrelationtoonlyonesubcategoryoftheothervariable.Forexample,inTable16.3figuresinparenthesesmarkedwith@aretherowpercentagescalculatedoutofthetotal(16)ofallagesubcategoriesofthevariableageinrelationtoonlyonesubcategoryofthesecondvariable(i.e.thosewhoholdastronglyfavourableattitudetowardsuraniummining)–inotherwords,onesubcategoryofavariabledisplayedonthestubbyallthesubcategoriesofthevariabledisplayedonthecolumnheadingofatable.Outofthosewhoholdastronglyunfavourableattitudetowardsuraniummining,21.4percentareundertheageof25years,noneisabovetheageof55andthemajority(42.9percent)arebetween25and34yearsofage(Table16.3).Thisrowpercentagehasthusgivenyouthevariationintermsofageamongthosewhoholdastronglyunfavourableattitudetowardsuraniummining.Ithasshownhowthe56respondentswhoholdastronglyunfavourableattitudetowardsuraniumminingdifferinagefromoneanother.Similarly,youcanselectanyothersubcategoryofthevariable(attitudetowardsuraniummining)toexamineitsvariationinrelationtotheothervariable,age.Columnpercentage–Inthesameway,youcanholdageataconstantlevelandexaminevariationsinattitude.Forexample,supposeyouwanttofindoutdifferencesinattitudeamong25–34-year-oldstowardsuraniummining.Theagecategory25–34(column)showsthatofthe36respondents,24(66.7percent)holdastronglyunfavourableattitudewhileonlytwo(5.5percent)holdastronglyfavourableattitudetowardsuraniummining.Youcandothesamebytakinganysubcategoryofthevariableage,toexaminedifferenceswithrespecttothedifferentsubcategoriesoftheothervariable(attitudestowardsuraniummining).Totalpercentage–Thisstandardisesthemagnitudeofeachcell;thatis,itgivesthepercentageofrespondentswhoareclassifiedinthesubcategoriesofonevariableinrelationtothesubcategoriesoftheothervariable.Forexample,whatpercentagedothosewhoareundertheageof25years,andholdastronglyunfavourableattitudetowardsuraniummining,constituteofthetotalpopulation?
Itispossibletosortdataforthreevariables.Table16.4(percentagesnotshown)examinesrespondents’attitudes in relation to their age and gender.As you addmore variables to a table it becomesmorecomplicatedtoreadandmoredifficulttointerpret,buttheprocedureforinterpretingitisthesame.The introduction of the third variable, gender, helps you to find out how the observed association
betweenthetwosubcategoriesofthetwovariables,ageandattitude,isdistributedinrelationtogender.Inotherwords,ithelpsyoutofindouthowmanymalesandfemalesconstituteaparticularcellshowingtheassociationbetweentheothertwovariables.Forexample,Table16.4showsthatofthosewhohavea stronglyunfavourable attitude towardsuraniummining,24 (42.9per cent) are25–34yearsof age.Thisgroupcomprises17(70.8percent)femalesand7(29.2percent)males.Hence, thetableshows
thatagreaterproportionoffemalethanmalerespondentsbetweentheagesof25and34holdastronglyunfavourable attitude towardsuraniummining.Similarly, you can take any two subcategories of ageandattitudeandrelatethesetoeithersubcategory(male/female)ofthethirdvariable,gender.
Graphs
Graphicpresentationsconstitutethethirdwayofcommunicatinganalyseddata.Graphicpresentationscanmakeanalyseddataeasiertounderstandandeffectivelycommunicatewhatitissupposedtoshow.Oneofthechoicesyouneedtomakeiswhetherasetofinformationisbestpresentedasatable,agraphor as text. Themain objective of a graph is to present data in away that is easy to understand andinterpret, and interesting to look at. Your decision to use a graph should be based mainly on thisconsideration:‘Agraphisbasedentirelyonthetableddataandthereforecantellnostorythatcannotbelearntbyinspectingatable.However,graphicrepresentationoftenmakesiteasiertoseethepertinentfeaturesofasetofdata’(Minium1978:45).Graphscanbeconstructedforeverytypeofdata–quantitativeandqualitative–andforanytypeof
variable(measuredonanominal,ordinal,intervalorratioscale).Therearedifferenttypesofgraph,andyourdecisiontouseaparticulartypeshouldbemadeonthebasisofthemeasurementscaleusedinthemeasurementofavariable.Itisequallyimportanttokeepinmindthemeasurementscalewhenitcomesto interpretation. It is not uncommon to find people misinterpreting a graph and drawing wrongconclusionssimplybecausetheyhaveoverlookedthemeasurementscaleusedinthemeasurementofavariable. The type of graph you choose depends upon the type of data you are displaying. Forcategoricalvariablesyoucanconstructonlybarcharts,histogramsorpiecharts,whereasforcontinuousvariables,inadditiontotheabove,lineortrendgraphscanalsobeconstructed.Thenumberofvariablesshowninagrapharealsoimportantindeterminingthetypeofgraphyoucanconstruct.Whenconstructingagraphofanytypeitisimportanttobeacquaintedwiththefollowingpoints:
Agraphicpresentationisconstructedinrelationtotwoaxes:horizontalandvertical.Thehorizontalaxisiscalledthe‘abscissa’or,morecommonly,thex-axis,andtheverticalaxisiscalledthe‘ordinate’or,morecommonly,they-axis(Minium1978:45).Ifagraphisdesignedtodisplayonlyonevariable,itiscustomary,butnotessential,torepresentthesubcategoriesofthevariablealongthex-axisandthefrequencyorcountofthatsubcategoryalongthey-axis.Thepointwheretheaxesintersectisconsideredasthezeropointforthey-axis.Whenagraphpresentstwovariables,oneisdisplayedoneachaxisandthepointwheretheyintersectisconsideredasthestartingorzeropoint.Agraph,likeatable,shouldhaveatitlethatdescribesitscontents.Theaxesshouldbelabelledalso.Agraphshouldbedrawntoanappropriatescale.Itisimportanttochooseascalethatenablesyourgraphtobeneithertoosmallnortoolarge,andyourchoiceofscaleforeachaxisshouldresultinthespreadofaxesbeingroughlyproportionatetooneanother.Sometimes,tofitthespreadofthescale(whenitistoospreadout)ononeorbothaxes,itisnecessarytobreakthescaleandalertreadersbyintroducingabreak(usuallytwoslantingparallellines)intheaxes.
Thehistogram
Ahistogram consists of a series of rectangles drawn next to each otherwithout any space between
them,eachrepresentingthefrequencyofacategoryorsubcategory(Figures,16.2a,b,c).Theirheightisinproportiontothefrequencytheyrepresent.Theheightoftherectanglesmayrepresenttheabsoluteorproportionalfrequencyorthepercentageofthetotal.Asmentioned,ahistogramcanbedrawnforbothcategorical and continuous variables. When interpreting a histogram you need to take into accountwhether it is representing categorical or continuous variables. Figures 16.2a, b and c provide threeexamplesofhistogramsusingdatafromTables16.1and16.4.Thesecondhistogramiseffectivelythesameasthefirstbutispresentedinathree-dimensionalstyle.
Thebarchart
Thebarchartordiagramisusedfordisplayingcategoricaldata(Figure16.3).Abarchartisidenticaltoahistogram,exceptthatinabarcharttherectanglesrepresentingthevariousfrequenciesarespaced,thusindicatingthatthedataiscategorical.Thebarchartisusedforvariablesmeasuredonnominalorordinal scales. The discrete categories are usually displayed along the x-axis and the number orpercentageof respondentson they-axis.However, as illustrated, it is possible todisplay thediscretecategoriesalongthey-axis.Thebarchart isaneffectivewayofvisuallydisplayingthemagnitudeofeachsubcategoryofavariable.
FIGURE16.2aTwo-dimensionalhistogram
FIGURE16.2bThree-dimensionalhistogram
FIGURE16.2cTwo-dimensionalhistogramwithtwovariables
FIGURE16.3Barcharts
Thestackedbarchart
Astackedbarchartissimilartoabarchartexceptthatintheformereachbarshowsinformationabouttwoormorevariablesstackedontoeachothervertically(Figure16.4).Thesectionsofabarshowtheproportionofthevariablestheyrepresentinrelationtooneanother.Thestackedbarscanbedrawnonlyforcategoricaldata.
FIGURE16.4Thestackedbarchart
The100percentbarchart
The100 per cent bar chart (Figure 16.5) is very similar to the stacked bar chart. In this case, thesubcategoriesofavariableareconvertedintopercentagesofthetotalpopulation.Eachbar,whichtotals100,isslicedintoportionsrelativetothepercentageofeachsubcategoryofthevariable.
FIGURE16.5The100percentbarchart
Thefrequencypolygon
The frequencypolygon isvery similar toahistogram.A frequencypolygon isdrawnby joining themidpointofeachrectangleataheightcommensuratewiththefrequencyofthatinterval(Figure16.6).One problem in constructing a frequency polygon is what to do with the two categories at eitherextreme.Tobringthepolygonlinebacktothex-axis,imaginethatthetwoextremecategorieshaveanintervalsimilartotherestandassumethefrequencyinthesecategoriestobezero.Fromthemidpointoftheseintervals,youextendthepolygonlinetomeetthex-axisatbothends.Afrequencypolygoncanbedrawnusingeitherabsoluteorproportionatefrequencies.
Thecumulativefrequencypolygon
Thecumulativefrequencypolygonorcumulativefrequencycurve(Figure16.7)isdrawnonthebasisof cumulative frequencies. The main difference between a frequency polygon and a cumulativefrequencypolygon is that the former is drawnby joining themidpointsof the intervals,whereas thelatterisdrawnbyjoiningtheendpointsoftheintervalsbecausecumulativefrequenciesinterpretdatainrelationtotheupperlimitofaninterval.Asacumulativefrequencydistributiontellsyouthenumberofobservations less thanagivenvalueand isusuallybasedupongroupeddata, to interpreta frequencydistributiontheupperlimitneedstobetaken.
FIGURE16.6Thefrequencypolygon
FIGURE16.7Thecumulativefrequencypolygon
Thestem-and-leafdisplay
Thestem-and-leafdisplayisaneffective,quickandsimplewayofdisplayingafrequencydistribution(Figure16.8).Thestem-and-leafdiagramforafrequencydistributionrunningintotwodigitsisplottedbydisplayingdigits0 to9on the leftof they-axis, representing the tensofa frequency.The figuresrepresentingtheunitsofafrequency(i.e.theright-handfigureofatwo-digitfrequency)aredisplayedon therightof they-axis.Note that thestem-and-leafdisplaydoesnotusegroupeddatabutabsolutefrequencies.Ifthedisplayisrotated90degreesinananti-clockwisedirection,iteffectivelybecomesahistogram.Withthistechniquesomeofthedescriptivestatisticsrelatingtothefrequencydistribution,suchasthemean,themodeandthemedian,caneasilybeascertained;however,theprocedurefortheircalculation isbeyond thescopeof thisbook.Stem-and-leafdisplaysarealsopossible for frequenciesrunningintothreeandfourdigits(hundredsandthousands).
FIGURE16.8Thestem-and-leafdisplay
Thepiechart
Thepiechartisanotherwayofrepresentingdatagraphically(Figure16.9),thistimeasacircle.Thereare 360 degrees in a circle, and so the full circle can be used to represent 100 per cent, or the totalpopulation. The circle or pie is divided into sections in accordance with the magnitude of eachsubcategory, and so each slice is in proportion to the size of each subcategory of a frequencydistribution.Theproportionsmaybeshowneitherasabsolutenumbersoraspercentages.Manually,piechartsaremoredifficult todraw thanother typesofgraphbecauseof thedifficulty inmeasuring thedegrees of the pie/circle. They can be drawn for both qualitative data and variables measured on acontinuousscalebutgroupedintocategories.
FIGURE16.9Two-andthree-dimensionalpiecharts
Thelinediagramortrendcurve
Asetofdatameasuredonacontinuousintervaloraratioscalecanbedisplayedusingalinediagramortrendcurve.A trend line canbe drawn for data pertaining to both a specific time (e.g. 1995, 1996,1997)oraperiod(e.g.1985–1989,1990–1994,1995–). If it relates toaperiod, themidpointofeachinterval at a height commensuratewith each frequency – as in the case of a frequency polygon– ismarkedasadot.Thesedotsarethenconnectedwithstraightlinestoexaminetrendsinaphenomenon.Ifthedatapertainstoexacttime,apointisplottedataheightcommensuratewiththefrequency.Thesepointsarethenconnectedwithstraightlines.Alinediagramisausefulwayofvisuallyconveyingthechangeswhenlong-termtrendsinaphenomenonorsituationneedtobestudied,orthechangesinthesubcategoryofavariablearemeasuredonanintervaloraratioscale(Figure16.10).Trendsplottedasalinediagramaremoreclearlyvisiblethaninatable.Forexample,alinediagramwouldbeusefulforillustratingtrendsinbirthordeathratesandchangesinpopulationsize.
Theareachart
Forvariablesmeasuredonanintervaloraratioscale,informationaboutthesubcategoriesofavariablecanalsobepresentedintheformofanareachart.Thisisplottedinthesamewayasalinediagrambutwiththeareaundereachlineshadedtohighlightthetotalmagnitudeofthesubcategoryinrelationtoothersubcategories.Forexample,Figure16.11showsthenumberofmaleandfemalerespondentsbyage.
FIGURE16.10Thelinediagramortrendcurve
FIGURE16.11Theareachart
Thescattergram
Whenyouwanttoshowvisuallyhowonevariablechangesinrelationtoachangeintheothervariable,ascattergramisextremelyeffective.
Forascattergram,boththevariablesmustbemeasuredeitheronintervalorratioscalesandthedataonboththevariablesneedstobeavailableinabsolutevaluesforeachobservation–youcannotdevelopascattergramforcategoricalvariables.Dataforbothvariablesistakeninpairsanddisplayedasdotsinrelation to theirvaluesonbothaxes.Letus take thedataonageand incomefor10respondentsofahypothetical study inTable16.5. The relationship between age and income based upon hypotheticaldataisshowninFigure16.12.
TABLE16.5Ageandincomedata
FIGURE16.12Thescattergram
Statisticalmeasures
Statisticalmeasures are extremely effective in communicating the findings in a precise and succinctmanner.Theiruseincertainsituationsisdesirableandinsomeitisessential,however,youcanconductaperfectlyvalidstudywithoutusinganystatisticalmeasure.Therearemanystatisticalmeasuresrangingfromverysimpletoextremelycomplicated.Ononeend
ofthespectrumyouhavesimpledescriptivemeasuressuchasmean,mode,medianand,ontheother;there are inferential statistical measures like analysis of variance, factorial analysis, multipleregressions.Becauseofitsvastness,statisticsisconsideredaseparateacademicdisciplineandbeforeyouareable
tousethesemeasures,youneedtolearnaboutthem.Useofstatisticalmeasuresisdependentuponthetypeofdatacollected,yourknowledgeofstatistics,
thepurposeofcommunicatingthefindings,andtheknowledgebaseinstatisticsofyourreadership.
Before using statistical measures, make sure the data lends itself to the application of statisticalmeasures,youhavesufficientknowledgeaboutthem,andyourreadershipcanunderstandthem.
SummaryResearchfindingsinbothquantitativeandqualitativeresearchareusuallyconveyedtoreadersthroughtext.Inqualitativeresearchthisismoreorlessthesolemethodofcommunication.However,inquantitativestudies,thoughtextisstillthedominantmethodofcommunicating research findings, it isoftencombinedwithother formssuchas tables,graphsandstatisticalmeasures.Thesecanmakecommunicationbetter,clearer,moreeffectiveandeasiertounderstand.Whatyouuseshouldbedeterminedbywhatyoufeelcomfortablewith,whatyouthinkwillbeeasiestforreaderstounderstandandwhatyouthinkwillenhancetheunderstandingofyourwriting.Tableshavetheadvantageofcontainingagreatdealofinformationinasmallspace,whilegraphsmakeiteasyforreaderstoabsorbinformationataglance.Usually,atablewillhavefiveparts:title,stub,columnheadings,bodyandsupplementarynotesorfootnotes.Dependinguponthe
number of variables aboutwhich information in a table is stored, there are three types of table: univariate (frequency), bivariate(cross-tabulation)andpolyvariate.Tointerpretatable,simplearithmeticproceduressuchaspercentages,cumulativefrequenciesorratioscanbeused.Youcanalso
calculate simple descriptive statistical procedures such as themean, themode, themedian, the chi-square test, the t-test and thecoefficientofcorrelation.Ifyouhavestatisticalknowledge,advancedstatisticscanbeapplied.Whiletherearemanytypesofgraphs,thecommononesare:thehistogram,thebardiagram,the
stackedbarchart,the100percentbarchart,thefrequencypolygon,thestem-and-leafdisplay,thepiechart,thelineortrenddiagram,theareachartandthescattergram.Whichisuseddependsuponyour purpose and themeasurement scale used tomeasure the variable(s) being displayed. Somegraphsaredifficulttodrawbutseveralcomputerprogramsarecapableofthis.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Identifytwospecificexampleswhereyoucoulduseatableratherthanjusttexttocommunicatefindingsandtwoexampleswheregraphswouldbebetter.Constructahypotheticalbivariatetable,withinthecontextofanareaofinterest.Calculatedifferenttypesofpercentagesandinterpretthedata.
STEPVIIIWritingaResearchReport
Thisoperationalstepincludesonechapter:
Chapter17:Writingaresearchreport
CHAPTER17WritingaResearchReport
Inthischapteryouwilllearnabout:
HowtowritearesearchreportHowtodevelopanoutlineforyourresearchreportWritingaboutavariableDifferentreferencingsystemsHowtowriteabibliography
Keywords: association, bibliography, intellectual rigour, non-spurious, outline,referencing,spurious,variable,verifiability.
Writingaresearchreport
Thelaststepintheresearchprocessiswritingtheresearchreport.Eachstepoftheprocessisimportantforavalidstudy,asnegligenceatanystagewillaffect thequalityofnot just thatpartbut thewholestudy.Inaway,thislaststepisthemostcrucialasitisthroughthereportthatthefindingsofthestudyandtheirimplicationsarecommunicatedtoyoursupervisorandreaders.Mostpeoplewillnotbeawareoftheamountandqualityofworkthathasgoneintoyourstudy.Whilemuchhardworkandcaremayhave been put into every stage of the research, all readers see is the report. Therefore, the wholeenterprise can be spoiled if the report is notwellwritten.AsBurnswrites, ‘extremely valuable andinterestingpracticalworkmaybespoiledatthelastminutebyastudentwhoisnotabletocommunicatetheresultseasily’(1997:229).Inaddition toyourunderstandingof researchmethodology, thequalityof the reportdependsupon
suchthingsasyourwrittencommunicationskillsandclarityofthought,yourabilitytoexpressthoughtsin a logical and sequentialmanner, and your knowledge base of the subject area.Another importantdeterminantisyourexperienceinresearchwriting:themoreexperienceyouacquire,themoreeffectiveyouwillbecomeinwritingaresearchreport.Theuseofstatisticalprocedureswillreinforcethevalidityofyourconclusionsandargumentsastheyenableyoutoestablishifanobservedassociationisduetochanceorotherwise(i.e.whetherarelationshipisspuriousornon-spurious)andindicatethestrengthofan association so readers can place confidence in your findings. The use of graphs to present the
findings,thoughnotessential,willmaketheinformationmoreeasilyunderstoodbyreaders.Asstatedinthepreviouschapter,whetherornotgraphsareuseddependsuponthepurposeforwhichthefindingsaretobeused.Themaindifferencebetweenresearchandotherwritingisinthedegreeofcontrol,rigorousnessand
cautionrequired.Researchwritingiscontrolledinthesensethatyouneedtobeextremelycarefulaboutwhatyouwrite,thewordsyouchoose,thewayideasareexpressed,andthevalidityandverifiabilityofthebasesfortheconclusionsyoudraw.Whatmostdistinguishesresearchwritingfromotherwritingisthehighdegreeofintellectualrigourrequired.Researchwritingmustbeabsolutelyaccurate,clear,freeofambiguity,logicalandconcise.Yourwritingshouldnotbebaseduponassumptionsaboutknowledgeofyourreadersaboutthestudy.Bearinmindthatyoumustbeabletodefendwhateveryouwriteshouldanyonechallengeit.Donotuseornamentalandsuperficiallanguage.Eventhebestresearchersmakeanumberofdraftsbeforewritinguptheirfinalone,sobepreparedtoundertakethistask.Theway findings are communicateddiffers in quantitative andqualitative research.Asmentioned
earlier,inqualitativeresearchthefindingsaremostlycommunicatedindescriptiveornarrativeformatwritten around the major themes, events or discourses that emerge from your findings. The mainpurpose is to describe the variation in a phenomenon, situation, event or episodewithoutmaking anattempttoquantifythevariation.OneofthewaysofwritingaqualitativereportisdescribedinChapter15asapartofthecontentanalysisprocess.Ontheotherhand,thewritinginquantitativeresearch,inadditiontobeingdescriptive,alsoincludesitsquantification.Dependinguponthepurposeofthestudy,statisticalmeasuresandtestscanalsobecomeapartoftheresearchwritingtosupportthefindings.
Developinganoutline
Beforeyou startwritingyour report, it isgoodpractice todevelopanoutline (‘chapterisation’).Thismeansdecidinghowyouaregoingtodivideyourreportintodifferentchaptersandplanningwhatwillbe written in each one. In developing chapterisation, the subobjectives of your study or the majorsignificant themes that emerged from content analysis can provide immense guidance. Develop thechaptersaroundthesignificantsubobjectivesorthemesofyourstudy.Dependingupontheimportanceofathemeorasubobjective,eitherdevoteacompletechaptertoitorcombineitwithrelatedthemestoformonechapter.Thetitleofeachchaptershouldbedescriptiveof themaintheme,communicate itsmainthrustandbeclearandconcise.Thisisapplicabletobothtypesofresearch.Thefollowingapproachisapplicabletobothqualitativeandquantitativetypesofresearchbutkeepin
mind that it ismerelysuggestiveandmaybeofhelp ifyouhaveno ideawhere tostart.Feel free tochangethesuggestedformatinanywayyoulikeorifyoupreferadifferentone,followthat.Thefirstchapterofyourreport,possiblyentitled‘Introduction’,shouldbeageneralintroductionto
the study, covering most of your project proposal and pointing out the deviations, if any, from theoriginal plan.This chapter covers all the preparatory tasks undertaken prior to conducting the study,suchas the literature review, the theoretical framework, theobjectivesof the study, studydesign, thesamplingstrategyandthemeasurementprocedures.Toillustratethis, twoexamplesareprovidedbelowforprojectsreferredtopreviouslyinthisbook:
thestudyonfoster-carepaymentsandtheFamilyEngagementmodel.Thefirstchaptersofthesereportscould be written around the subheadings below. The subsequent structure of these reports is quitedifferent.Keeping inview thepurpose forwhichFamilyEngagement evaluationwas commissioned,the report was divided into three parts: the Introduction, the perceived model, and conclusions andrecommendation.
Attitudestowardsfoster-carepayments:suggestedcontentsofchapter1Chapter1Introduction
IntroductionThedevelopmentoffostercareFostercareinAustraliaFostercareinWesternAustraliaTheDepartmentofCommunityServicesTheout-of-homeandcommunitycareprogrammeCurrenttrendsinfoster-careplacementinWesternAustraliaBecomingafostercarerFoster-caresubsidiesIssuesregardingfoster-carepaymentRationaleforthestudyObjectivesofthestudyStudydesignSamplingMeasurementprocedureProblemsandlimitationsWorkingdefinitions
TheFamilyEngagement–Aservicedeliverymodel:suggestedcontentsofchapter1PartOne:Introduction
Background:TheoriginoftheFamilyEngagementideaHistoricalperspectiveTheperceivedmodel
ConceptualframeworkPhilosophicalperspectiveunderpinningthemodelIndentedoutcomes
ObjectivesoftheevaluationEvaluationmethodology
(Note: In this section, the conceptual framework of themodel, its philosophical basis, perceived outcomes as identified by theperson(s)responsibleforinitiatingtheidea,andwhatwasavailableintheliterature,wereincluded.Italsoincludeddetailsaboutevaluationobjectivesandevaluationmethodology.)
The second chapter in quantitative research reports should provide information about the studypopulation.Here,therelevantsocial,economicanddemographiccharacteristicsofthestudypopulationshouldbedescribed.Thischapterservestwopurposes:
1. Itprovidesreaderswithsomebackgroundinformationaboutthepopulationfromwhichyoucollectedtheinformationsotheycanrelatethefindingstothetypeofpopulationstudied.
2. Ithelpstoidentifythevariancewithinagroup;forexample,youmaywanttoexaminehowthelevelofsatisfactionoftheconsumersofaservicechangeswiththeirage,genderoreducation.
The second chapter in a quantitative research report, therefore, could be entitled ‘Socioeconomic–demographiccharacteristicsofthestudypopulation’orjust‘Thestudypopulation’.Thischaptercouldbewrittenaroundthesubheadingsbelowwhichareillustratedbytakingtheexampleofthefoster-carepaymentstudy.Asqualitativestudiesaremostlybaseduponalimitednumberofin-depthinterviewsorobservations,
youmayfinditverydifficulttowriteaboutthestudypopulation.
Attitudetowardsfoster–carepayments:suggestedcontentsofchapterIIChapterIIThestudypopulation
IntroductionRespondentsbyage(Informationobtainedinresponsetothequestiononageshouldbepresentedhere.Consult‘Writingaboutavariable’,thenextsectionofthischapter.)Respondentsbygender(Followthesuggestionsmadeunder‘Writingaboutavariable’(seebelow)fortherestofthevariables.)MaritalstatusofthestudypopulationEthnicityofrespondentsStudypopulationbynumberofchildrenAnnualaverageincomeofthestudypopulationStudypopulationbytypeofdwellingetc.
The title and contents of subsequent chapters depend uponwhat you have attempted to describe,explore, examine, establish or prove in your study.As the content of each project is different, thesechapterswillbedifferent.Asindicatedearlier,thetitleofeachchaptershouldreflectthemainthrustofitscontents.The outline should specify the subsections of the chapter. These subsections should be developed
around the different aspects of the themebeingdiscussed in the chapter. If youplan to correlate theinformation obtained from one variable with another, specify the variables. Plan the sequence fordiscussionofthevariables.Indecidingthis,keepinmindthelinkageandlogicalprogressionbetweenthesections.Thisdoesnotmeanthattheproposedoutlinecannotbechangedwhenwritingthereport–itispossibleforittobesignificantlychanged.However,anoutline,evenifextremelyrough,willbeofimmensehelptoyou.Again,letustakethestudyonfoster-carepaymentandtheFamilyEngagementmodelasexamples:
Attitudestowardsfoster-carepayments:suggestedcontentsofchapterIIIChapterIIIAttitudestowardsthepresentlevelofpaymentforfostercare
IntroductionAttitudestowardsadequacyofpaymentforfostercare(Responsestoquestionsonthe
adequacyoffoster-carepaymentshouldbepresentedhere.)
adequacybyage(Cross-tabulation,i.e.responsestothequestiononadequacyoffoster-carepayment,isexaminedinrelationtotheresponsestoquestionsonage.)adequacybymaritalstatus(Cross-tabulation,i.e.responsestothequestiononadequacyoffoster-carepayment,isexaminedinrelationtotheresponsestoquestionsonmaritalstatus.)adequacybyincomeofthefamily(Cross-tabulation,i.e.responsestothequestiononadequacyoffoster-carepayment,isexaminedinrelationtotheresponsestoquestionsonincome.)
AspectsoffostercarenotcoveredbythepaymentMajorcostsbornebyfostercarersEffectsofthecurrentlevelofpaymentonthefamilyReasonsforincreasingthepaymentProposedlevelofpayment
proposedlevelbyincomeofthefamily
Conclusions
(Note:Cross-tabulationscanbeincludedforanyvariablewhereappropriate.)
FamilyEngagementmodel:suggestedcontentsofchapterIIPartTwo:Theperceivedmodel
ThephilosophyunderpinningthemodelDevelopmentofthemodelThemodelinpracticePerceiveddifferencesinpracticebeforeandaftertheintroductionofthemodelPerceivedstrengthsofthemodelPerceivedweaknessesofthemodelSkillsrequiredforeffectivefunctioningunderthemodelReplicationofthemodelReasonsforchangetothenewmodelTraining
Howshouldstaffbetrained?Trainingprovided
NameofthemodelDeterminantsofsuccessfulimplementationofthemodelIndicatorsofsuccessofthemodelWhatcouldhavebeendonedifferently?Whatneedstobedonetoimprovethemodel?
RoleofCommunityDevelopmentFundingOfficersAdvantagesanddisadvantagesoftheCaseManagementmodelSatisfactionofstaffwiththemodelThemodelanddepartmentalvision,philosophy,ethos,principlesAttitudeofclientstowardsthemodelAttitudeofcommunityagenciestowardsthemodelThemodelandchangesintheselectedindicators
(Note: In this section, findings about different aspects of the model as identified through in-depth interviews and focus groupdiscussionsweredetailed.)
FamilyEngagementmodel:suggestedcontentsofchapterIII
PartThree:Conclusionsandrecommendations
Conclusions
A:GeneralB:Specifictothemodel
Recommendations
A:GeneralB:Specifictothe…office
Thistypeofoutlineprovidesdirectioninyourwriting.Asmentionedearlier,asyoustartwritingyouwillcertainlychangeit,butneverthelessyouwillfinditveryhelpfulinyourwrite-up.
Writingaboutavariable
Having developed a chapter outline, the next step is to start writing. Though the way researchersorganise their writing is extremely individualised, the following guidelines and format may provehelpfulforbeginners.Whenwriting about the information obtained in response to a question (variable),write as if you
wereprovidinganswerstothefollowingquestions:
Whydidyouthinkitimportanttostudythevariable?Whateffects,inyouropinion,maythisvariablehaveonthemainvariableyouareexplaining?(Thisiswhereyouprovideyourownrationaleforstudyingthevariable.)Inthecaseofacross-tabulation,whatrelationshipshaveotherstudiesfoundbetweenthevariablesyouareanalysing?(Thisiswheretheliteraturereviewisintegratedintothefindingsofthestudy.)Whatdidyouexpecttofindoutintermsoftherelationshipbetweenthetwovariables?(Ifyou
haveformulatedahypothesis,stateithere.)Whathasyourstudyfoundout?(Providetheharddatafromyourstudyhere,astables,graphsortext.)Whatdoesthedatashow?(Interpretthefindingsofyouranalysis.)Whatconclusionscanyoudraw?Howdotheconclusionsdrawnfromyourstudycomparewiththosefromsimilarstudiesinthepast?Doesyourstudysupportorcontradictthem?Whatexplanationcanyouprovideforthefindingsofyourstudy?
Theaboveisonlyasuggestedformatfororderingyourthoughts,notalistofsubheadings.Youmaywish to change the suggested order to make the reading more interesting. Below is an example ofwritingaboutavariable,‘Adequacyofpaymentforfostercare’,fromChapter13:
Whydidyouthinkitimportanttofindoutiffoster-carepaymentsareadequate?Whateffects,inyouropinion,couldtheadequacyorotherwiseofpaymentforfostercarehaveonthequalityoffostercare?Whathaveotherstudiesinyourliteraturereviewsaidabouttheadequacyoffoster-carepayments?Whatdidyouexpecttofindoutfromyourstudypopulationintermsofifitsfeelingsabouttheadequacyoffoster-carepayments?Ifyouformulatedahypothesis,youshouldspecifythathere.Forexample,Hi=Mostfosterparentswouldconsiderthecurrentleveloffoster-carepaymentstobeadequate.Whatdidyoufindoutabouttheadequacyoffoster-carepayments?Whatproportionofthestudypopulationsaidtheywereadequate?Whatproportionsaidtheywereinadequate?Provideatableorgraphshowingthedistributionofrespondentsbytheirresponsetothequestionregardingtheadequacyoffoster-carepayments.Whatdoesyourdatashowabouttheadequacyoffoster-carepayments?Whatarethemainfindingsofyourstudy?Howdothesefindingscomparewiththoseofotherstudiesyoufoundinyourliteraturereview?Doesyourstudysupportorcontradictthem?Whatconclusionscanyoudrawabouttheadequacyoftheamountofpaymentforfostercare?Whatexplanationcanyouprovidefortheobservedfindings?Whydoyouthinkthosewhosaidthatfosterpaymentsareeitheradequateorinadequatefeelthatway?
In the suggested format in writing about information obtained from questions, notice that theliterature review is integratedwith the findings and conclusions.The extent of the integrationof theliterature with findings mostly depends upon the level at which you are writing your dissertation(Honours,MastersorPhD)–thehigherthelevel,themoreextensivetheliteraturereview,thegreateritsintegration with your findings, and the more careful and confident you need to be about yourconclusions.Writing inqualitative research ismoredescriptiveandnarrative thananalytical,henceyouneed to
useyourimaginationintermsofplacementofinformation,linkagebetweenthethoughtsandflowoflanguagetomakethewritinginterestingtoreadandmeaningfulinconveyingthefindings.The suggested format is organised around themain themes of the study. There are other formats.
Someresearcherswriteeverythingunderoneheading,‘Thefindings’.Thisformatisappropriateforaresearchpaper,becauseitisshort,butnotforaresearchreportordissertation.Otherwritersfollowthesameorderasintheresearchinstrument;forexample,findingsarediscussedundereachquestion.Thereaderneedstorefercontinuouslytotheinstrumentforeachquestion.Itissegmental,lackslinkageandintegration,anddoesnotplacefindingsintoperspective.
Referencing
Thereportshouldfollowanacademicstyleofreferencing.AccordingtoButcher(1981:226),therearefourreferencingsystemsfromwhichtochoose:
1. theshort-titlesystem;2. theauthor–datesystem;3. thereferencebynumbersystem;4. theauthor–numbersystem.
Youneed toadopt theone that isacceptable toyouruniversityandacademicdiscipline: ‘Thefirstoftheseisusedinmostgeneralbooks,thesecondmainlyinscienceandsocialsciencebooks;thethirdandfourthlessfrequently’(Butcher1981:167).
Writingabibliography
Again, there are several well-established systems for writing a bibliography and your choice isdependentuponthepreferenceofthedisciplineanduniversity.Inthesocialsciencessomeofthemostcommonlyusedonesare(Longyear1983:83):
theHarvardsystem;theAmericanPsychologicalAssociationsystem;theAmericanMedicalAssociationsystem;theMcGraw-Hillsystem;theModernLanguagesAssociationsystem;thefootnotesystem.
To learnabout thesesystemsandstyles,consult the referencesprovidedat theendof thisbookorconsultyourlibrary.
SummaryIn a way, writing your report is the most crucial step in the research process as it communicates the findings to your researchsupervisorandreaders.Abadlywrittenreportcanspoilallthehardworkyouhaveputintoyourresearchstudy.Styles of research writing vary markedly among researchers but all research reports must be written clearly and concisely.
Furthermore, scientific writing requires intellectual rigour and there are certain obligations in terms of accuracy and objectivity.Reportscanbewrittenindifferentformatsandthischapterhassuggestedonethatresearchstudentshavefoundtobehelpful.Writing inquantitative andqualitative researchdiffers to the extent that in qualitative research
your style is descriptive and narrative, whereas in quantitative research, in addition to beingdescriptive, it is also analytical and every assertion is supported by empirical evidence gatheredthroughtheinvestigation.There are different ways of referencing and of writing a bibliography. You need to select the
systemthatisacceptabletoyourdisciplineanduniversity.Beforeyoustartwritingtheresearchreport,developanoutlineofthedifferentchaptersandtheir
contents. The chapters should bewritten around themain themes of the study and for this yoursubobjectivesareofimmensehelp.Whenprovidingspecificinformationaboutavariable,thewrite-upshould integrate therationale forstudying thevariable; the literature review; thehypothesis, ifany;findings;conclusionsdrawn;andpossibleexplanationsforthefindings.The suggested format can be described as thematicwriting – writing organised around the significant themes of your study.
Withinathemetheinformationisprovidedinanintegratedmannerfollowingalogicalprogressionofthought.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Aliteraturereviewisanintegralpartofresearchwriting.Reflectingonexamplesfromyourownareaofinterest,explorehowyoumightbeabletointegrateyourresearchfindingswithyourliteraturereviewwhenitcomestowritingyourreport.Canyouthinkofthreewaysinwhichreportwritinginqualitativeandquantitativeresearchdiffers?
CHAPTER18ResearchMethodologyandPracticeEvaluation
Inthischapteryouwilllearnabout:
WhatevaluationisandwhyitisdoneTheprocessforusingevaluationtodevelopaninterventionThetwodifferentperspectivesintheclassificationofevaluationstudiesTypesofevaluationfromafocusperspectiveTypesofevaluationfromaphilosophicalperspectiveTheprocessofundertakinganevaluationTheimportanceofinvolvingstakeholdersinevaluationEthicsinevaluation
Keywords: client-centred evaluation, cost–benefit evaluation, cost-effectivenessevaluation, ethics, evaluation, evaluation process, goal-centred, holistic evaluation,illuminative evaluation, impact evaluation, improvement-oriented evaluation,indicators, intervention, monitoring, objective-oriented evaluation, outcomeevaluation,perspectives,processevaluation,stakeholders.
Research methodology and practice evaluation are integrally related. Practice evaluation relies veryheavilyonthetechniques,methodsandskillsofresearchmethodology.Foranevaluatoritisimperativeto be a good researcher. As this book is primarily written for newcomers to research and forpractitioners in human services who are increasingly being asked to provide evidence of theeffectiveness of their practice, it is only appropriate that this book includes a chapter that brieflyoutlinesevaluationresearchanditsrelationshipwithresearchmethodology.Over the past few decades evaluation research has gained greater prominence and has developed
rapidly, in both its applications andmethodology. Scarcity of resources, emergence of a need to beaccountableforeffectiveandefficientdeliveryofservices,realisationthatconsumershavetherighttoknowaboutthequalityoftheservicetheyarereceiving,andtheonsetofaneraofeconomicrationalismhaveallcontributedtothisrapiddevelopment.Thoughitreliesveryheavilyonthecontentsofresearchmethodologyper se, evaluation research is nowconsidered to be a self-defineddiscipline in its ownright,withitsownliterature,techniquesandskills.Methodsandmodelsofevaluationhavenowbeenapplied to almost every field of knowledge in our society. Evaluators are being engaged to evaluate
manysocial,economic,health,educationandpoliticalprogrammes.Theveryfirstquestionthatmaycometoyourmind,asabeginner,is:whatisevaluationresearch?
Evaluation may have a different meaning in different situations and, also, it may be understooddifferentlybydifferentpeople.Itis,therefore,importantforyoutounderstandthevariousperspectiveson and aspects of evaluation, so that when you come upon it you can define its meaning for yoursituation.
Whatisevaluation?
Ifyougothroughtheliteratureonevaluationresearch,youwillcomeacrossmanydifferentdefinitions.Belowaresomedefinitionsthathavebeenselectedtohighlightthevariousdimensionsofevaluation.According toRossi,FreemanandLipsey(1999:4): ‘Programevaluation is theuseofsocial researchprocedurestosystematicallyinvestigatetheeffectivenessofsocialinterventionsprograms.’As quoted by Stufflebeam and Shinkfield (1985: 3), the definition of the Joint Committee on
StandardsforEducationEvaluationis:‘Evaluationisthesystematicassessmentoftheworthandmeritofsomeobjects.’AccordingtoAlkinandSolomon(1983:14):
Evaluation is a process of ascertaining the decision areas of concern, selecting appropriateinformation, andcollectingandanalysing information inorder to report summarydatauseful todecisionmakersinselectingamongalternatives.
AccordingtoRutman(1980:17),‘Programevaluationreferstotheuseofresearchmethodstomeasuretheeffectivenessofoperativeprograms.’Inanotherbook,editedbyRutman(1977:16),healsousesthefollowingdefinition:
Evaluation research is, first and foremost, a process of applying scientific procedures toaccumulate reliable andvalid evidence in themanner and the extent towhich specific activitiesproduceparticulareffectsoroutcomes.
If you critically examine these definitions, you will notice that in the evaluation process (as inresearchmethodology)therearecertainpropertiessuchasvalidity,reliabilityandthoroughness.
FIGURE18.1Theconceptofevaluation
And both processes are designed to collect and analyse information in order to answer researchquestions. In evaluation research, research questions mainly revolve around various aspects of anintervention,programmeorpractice,whereasingeneralresearchtheymayrelatetoanyaspectorissueof concern or significance. Evaluation research, therefore, is primarily concerned with a criticalexamination of such aspects as the appropriateness, effectiveness and efficiency of an intervention.Issuesrelatingtoefficiencyandeffectivenessinrelationtothecostsandbenefitsofaninterventionarealsoanintegralpartofevaluationstudies.SeeFigure18.1.
Whyevaluation?
Suppose you are working in a human service agency. At some point in the course of your work,questionsmaycometoyourmindabouttheappropriatenessofyourservice,itseffectiveness,whysomepeoplelikeorbenefitfromitandothersdonot,howitcanbeimproved,whatsortofworkloadaserviceprovidercancarryandwhatthecostofdeliveringtheserviceis.Consumersandadministratorsofyourservicemayaskyousimilarquestions.Youcanobtainanswerstothesequestionsinanumberofways,rangingfromgatheringanecdotalevidencetoundertakingasystematicstudy,adheringtotheprinciplesof scientific enquiry. Evaluation methodology, which (as mentioned) is based upon researchmethodology,isonewayoffindinganswerstosuchquestions.Youmaycomeacrossprofessionalswithdifferingattitudestowardsevaluation.Someattachimmense
importance to it,while others consider it to benot as important because they thinkof themselves assolely theprovidersofa service.Whetherornotyoubecome involved inevaluatingyourpractice isdependentuponyourinterestinexaminingthepracticeanduponthedemandsplacedonyoubyothers.However,asabeginnerinresearchmethodology,youneedtobeawareoftheimportanceofevaluationandofthelinksbetweenitandresearchmethodology.Also,youneedtoappreciatethesignificanceofevaluationincriticallyexaminingapracticeforgreaterefficiencyandeffectiveness.Evenasaserviceprovideryouneed tobe familiarwithhowyourclinical skillscanbenefit fromevaluationprocesses.Specifically:
Youhaveaprofessionalandethicalresponsibilitytoprovideagoodqualityofservicetoyourclients.Toensureitseffectivenessandefficiency,youneedtoassessyourpractice.Knowledgeofevaluationresearchwillhelpyoutoassessyourpracticeobjectivelyorhelpyoutocommunicatewithanevaluatorknowledgeablyandprofessionallyaboutevaluationissues.Whileyou,asaprofessional,haveanobligationtoprovideaneffectiveservicetoyourclients,yourclients,ontheotherhand,havearighttoknowthequalityoftheservicetheyarereceivingfromyou.Inthisageofconsumerism,yourclientscandemandevidenceofthequalityofyourservice.Inthemoderneraofconsumerism,theemphasisisnotonlyonprovidingaservicebutalsoonhowwellitisdeliveredtoconsumers.Inmostserviceprofessionstheconceptofso-calledevidence-basedpracticeisgrowingataveryrapidrate.(Seealsothesectiononevidence-basedpracticeinChapter1.)Whenyouaredependentuponoutsidefundingforprovidingaservice,youusuallyneedtoprovideevidenceoftheeffectivenessofyourserviceforrenewaloffunding.Nowadaysalmosteveryfundingbodyusesevaluationreportsasthebasisoffundingdecisions.Quiteoftenanevaluationreportfromanindependentevaluatorisrequired.Foreffectivecommunicationwithanoutsideevaluator,knowledgeofevaluationwillgoalongway.Becauseofthepaucityofresourcesandagreateremphasisoneconomicrationalismnowadays,thereisagrowingdemandonserviceproviderstodemonstratethattheservicetheyareprovidingisworththeexpenditure,andpeoplearegettingvalueformoney.Criticalexaminationthroughevaluationofyourservicewillhelpyoutodemonstratetheworthandvalueofyourservice.Howdoconsumersviewyourservice?Whatdotheconsumersofyourservicefeelaboutit?Whatdotheyseeasthepositiveaspectsofyourservice?What,intheiropinion,arethenegativeaspects?Howcanyourservicebeimproved?isyourservicereallyhelpingthoseforwhomitwasdesigned?Isitachievingitsobjectives?Inwhatwaysisitbenefitingyourclients?Toanswersuchquestionsyouneedtoevaluateyourpractice.Howexpensiveisyourservice?Whatisthecostofprovidingtheservicetoclients?Isthiscostjustified?Isthemoneybeingwellspent?
Inthefinaltwopointsabovearesomeofthequestionsthatyouneedtoanswerasaserviceprovider.Skillsinevaluationresearchcanhelpyoutoanswerthesequestionswithgreaterconfidence,objectivityandvalidity.
Intervention–development–evaluationprocess
Tounderstand theevaluationprocessforan intervention, it is important thatyoualsoknowhowit islinked to thedevelopmentof an intervention.The intervention–development–evaluationprocess isdividedintofourphases(Figure18.2):
1. needsassessment;2. intervention/programmedevelopment;3. intervention/programmeexecution;4. intervention/programmeevaluation.
FIGURE18.2Theintervention–development–evaluationmodelThedevelopmentofaninterventionusuallystartswithanassessmentoftheneedsofacommunity,
grouporpeoplelivinginageographicalarea(phase1).Basedupontheneeds,theaimsandobjectivesforaprogrammearedeveloped tomeet theseneeds,which in turnbecome thebasisofdevelopingaconceptual intervention programme. This conceptual construction is primarily based on previousexperiences, understanding of the problem area, knowledge about how others have dealt with theproblem in other places and/or opinionof experts in the area. In the development of this conceptualmodel, particular attention is given to the formulation of strategies to achieve the objectives of theprogramme.Next,thepreciseactivitiesneededtoachievethesestrategiesareidentified.Proceduresforundertakingtheseactivitiesarethendrawnup.Theseactivitiesandproceduresconstitutethecontentsofaprogramme(phase2).Ofcourse,theymayneedtobestreamlined,modifiedorotherwisechangedinthelightofexperience.Sometimes,aconceptual–interventionmodelisfirst‘tested’outasafeasibilitystudy to identifyproblemsandmodificationsbefore launchingon a full scale.Having fine-tuned theinterventioncontents,itisexecutedinaccordancewiththeproposedplan(phase3).Services/activitiesconstitute programme inputs, which result in intervention outputs, which in turn produceoutcomes/impacts.Outputsarethedirectproductsofaprogramme’sactivitiesandareusuallymeasuredin terms of volume of tasks accomplished. Outcomes are benefits or changes in individuals orpopulationsthatcanbeattributedtotheinputsofaprogramme.Theymaymanifestascognitiveand/ornon-cognitive changes. These may relate to values, attitudes, knowledge, behaviour, change in asituation or any other aspect that came about in an individual following the introduction of aprogramme.Thoughsomeevaluationsarefocusedontheprocessbywhichaserviceisdelivered(phase3),themajorityofevaluationsarearoundeitheroutputsoroutcomes(phase4).Let us take an example: randombreath testing (RBT). InRBT the outputs include the number of
peopletested;thenumberofawarenesscampaignsorganised;thenumberofnewspaperandtelevisionadvertisements placed; the number of community forums held; and the number of police officersemployed for the task of breath testing. The desired outcomes – the changes sought in people’sbehaviourandthesituation–mayincludeareductioninalcohol-relatedroadaccidentsanddeaths,andareductioninthenumberofpeoplecaughtdrivingundertheinfluenceofalcohol.Let us take another example: the counselling service for couples with marital problems. In this
example the outputs are the number of sessions with couples and the number of couples seen. Theoutcomesmightbeareductionintheconflicts;greatermaritalstabilitywithabeneficialeffectonthecouple’schildren;apositiveeffectonwork,productivityandincome;increasedsatisfactionwithlifeingeneral;orsmoothseparationbythecouplefromeachother.
Perspectivesintheclassificationofevaluationstudies
Thevarioustypesofevaluationcanbelookedatfromtwoperspectives:
thefocusoftheevaluation;thephilosophicalbasethatunderpinsanevaluation.
It is important to remember that these perspectives are not mutually exclusive. All evaluationscategorisedfromtheviewpointoffocusofevaluationhaveaphilosophicalbaseunderpinningthem,andsocanbeclassifiedfromwithinthisperspectiveaswell.Forexample,an impact/outcomeevaluation fromthefocus-of-evaluationpointofviewcanalsobe
classified as agoal-centredevaluation from the philosophical perspective. In an outcome evaluation(classifiedfromthefocus-of-evaluationperspective),youcaneitherexplorethewayaninterventionhasimpactedon the studypopulation, or seek to determineoutcomesby establishingwhether or not theprogramme has achieved its intended objective. If the evaluation is from the focus perspective, it isclassifiedasanimpact/outcomeevaluation,andifthefocusisfromthephilosophicalperspective,itisalso classified as a goal-centred evaluation. Again, if you determine the impact of aprogramme/intervention by askingwhat clients/consumers perceive its effects to have beenon them,this is alsoclassifiedas aclient-centredevaluation fromaphilosophical perspective. If you examineevery aspect of a programme with regard to its outcome, process and any other aspect, this iscategorised as a holistic evaluation. Finally, every type of evaluation, process or outcome can beclassifiedasanimprovement-orientedevaluationfromthephilosophicalperspectiveastheultimateaimof any evaluation is to improve an intervention/programme. To avoid confusion between the twoperspectives,anintegratedpictureisprovidedinFigure18.3.
FIGURE18.3Perspectivesintheclassificationofevaluationstudies
Typesofevaluationfromafocusperspective
From the perspective of the focus of evaluation there are four types of evaluation:programme/intervention planning, process/monitoring, impact/outcome and cost–benefit/cost-effectiveness. Each type addresses amain and significantly different issue.Evaluation forplanningaddresses the issue of establishing the need for a programme or intervention; process evaluationemphasisestheevaluationoftheprocessinordertoenhancetheefficiencyofthedeliverysystem;themeasurementofoutcomesisthefocusofanoutcomeevaluation;andthecentralaimofacost–benefitevaluationistoputapricetagonaninterventioninrelationtoitsbenefits.Hence,fromthisperspective,theclassificationofanevaluationisprimarilydependentuponitsfocus.It is important for you to understand the different evaluation questions that each is designed to
answer.Table18.1willhelpyoutounderstandtheapplicationofeachtypeofevaluation.
Evaluationforprogramme/interventionplanning
Inmany situations it is desirable to examine the feasibility of starting a programme/intervention byevaluatingthenatureandextentofachosenproblem.Actually,thistypeofstudyevaluatestheproblemperse:itsnature,extentanddistribution.Specifically,programmeplanningevaluationincludes:
estimatingtheextentoftheproblem–inotherwords,estimatinghowmanypeoplearelikelytoneedtheintervention;delineatingthecharacteristicsofthepeopleandgroupswhoarelikelytorequiretheintervention;identifyingthelikelybenefitstobederivedfromtheintervention;developingamethodofdeliveringtheintervention;developingprogrammecontents:services,activitiesandprocedures;identifyingtrainingneedsforservicedeliveryanddevelopingtrainingmaterial;estimatingthefinancialrequirementsoftheintervention;developingevaluationindicatorsforthesuccessorfailureoftheinterventionandfixingatimelineforevaluation.
Thereareanumberofmethodsforevaluatingtheextentandnatureofaproblem,andfordevisinga service deliverymanner. The choice of a particularmethod should depend upon the financialresourcesavailable,thetimeatyourdisposalandthelevelofaccuracyrequiredinyourestimates.Someofthemethodsare:
Communityneed-assessmentsurveys–Need-assessmentsurveysarequiteprevalenttodeterminetheextentofaproblem.Youuseyourresearchskillstoundertakeasurveyintherelevantcommunitytoascertainthenumberofpeoplewhowillrequireaparticularservice.Thenumberofpeoplerequiringaparticularservicecanbeextrapolatedusingdemographicinformationaboutthecommunityandresultsfromyourcommunitysamplesurvey.Ifdoneproperly,aneed-assessmentsurveycangiveyouareasonablyaccurateestimateoftheneedsofacommunityortheneedforaparticulartypeofservice.However,youmustkeepinmindthatsurveysarenotcheaptoundertake.Communityforums–Conductingcommunitydiscussionforumsisanothermethodusedtofindouttheextentoftheneedforaparticularservice.However,itisimportanttokeepinmindthatcommunityforumssufferfromaprobleminthatparticipantsareself-selected;hence,thepictureprovidedmaynotbeaccurate.Inacommunityforumnoteveryonewillparticipateandthosewhodomayhaveavestedinterestfororagainsttheservice.If,somehow,youcanmakesurethatall
interestgroupsarerepresentedinacommunityforum,itcanprovideareasonablepictureofthedemandforaservice.Communityforumsarecomparativelycheaptoundertakebutyouneedtoexaminetheusefulnessoftheinformationforyourpurpose.Withcommunityforumsyoucannotascertainthenumberofpeoplewhomayneedaparticularservice,butyoucangetsomeindicationofthedemandforaserviceanddifferentprevalentcommunityperspectiveswithrespecttotheservice.
TABLE18.1Typesofevaluationfromtheperspectiveofitsfocusandthequestionstheyaredesignedtoanswer
Socialindicators–Makinguseofsocialindicators,inconjunctionwithotherdemographicdata,ifyouhaveinformationaboutthem,isanothermethod.However,youhavetobecarefulthattheseindicatorshaveahighcorrelationwiththeproblem/needandareaccuratelyrecorded.Otherwise,theaccuracyoftheestimateswillbeaffected.Servicerecords–Therearetimeswhenyoumaybeabletouseexistingservicerecordstoidentifytheunmetneedsforaservice.Forexample,ifanagencyiskeepingarecordofthecaseswhereithasnotbeenabletoprovideaserviceforlackofresources,youmaybeabletouseittoestimatethenumberofpeoplewhoarelikelytoneedthatservice.Focusgroupsofpotentialserviceconsumers,serviceprovidersandexperts–Youcanalsousefocusgroupsmadeupofconsumers,serviceprovidersandexpertstoestablishtheneedforaservice.
Community surveys and social indicators tend to be quantitative, whereas the others tend to bequalitative.Thustheygiveyoudifferenttypesofinformation.Servicerecordsprovideanindicationofthegapinserviceandarenotreflectiveofitsneed.Itisimportanttorememberthatallthesemethods,exceptthecommunityneedssurvey,provideonly
anindicationofthedemandforaserviceinacommunity.Youhavetodeterminehowaccuratelyyouneedtoestimatethepotentialnumberofuserstostartaservice.Acommunitysurveywillprovideyouwiththemostaccuratefiguresbutitcouldputastrainontheresources.Also,keepinmindthatuseofmultiplemethodswillproducemoreaccurateestimates.
Process/monitoringevaluation
Processevaluation,alsoknownasmonitoringevaluation,focusesonthemannerofdeliveryofaserviceinordertoidentifyissuesandproblemsconcerningdelivery.Italsoidentifieswaysofimprovingservicedeliveryproceduresforabetterandmoreefficientservice.Specifically,processevaluationisusedfor:
determiningwhetherornotthedeliveryofaserviceisconsistentwiththeoriginaldesignspecificationsand,ifnot,foridentifyingthereasonsandjustificationsfornon-compliance;identifyingchangesneededinthedeliverymannerforgreatercoverageandefficiency;ascertaining,whenaninterventionhasnoimpact,whetherthisisbecauseoftheinterventionitselforthemannerinwhichitisbeingdelivered;determiningwhetherornotaninterventionisreachingtheappropriatetargetpopulation.
Processevaluationincludesevaluatingthe:
extentofparticipationofthetargetpopulation;deliverymannerofaprogramme/intervention.
FIGURE18.4AspectsofprocessevaluationEvaluating the participation of the target population in turn involves: (1) ascertaining the
appropriatenessoftheclientsfortheserviceinquestion;and(2)establishingthetotalnumberofclientsand thedropout rateamong them.Evaluating the servicedeliverymanner, in the sameway, includestwo tasks: (1) examining the procedures used in providing the service; and (2) examining the issuesrelatingtotheaccessibilityoftheservicetopotentialclients(Figure18.4).
Evaluatingparticipationofthetargetpopulation
Inanevaluationstudydesignedtoexaminetheprocessofdeliveringanintervention,itisimportanttoexaminetheappropriatenessoftheusersoftheservicebecause,sometimes,somepeopleuseaserviceeventhoughtheydonotstrictlyfallwithintheinclusioncriteria.Inotherwords,inevaluationstudiesitis important to determine not just the number of users, but whether or not they are eligible users.Determiningtheappropriateuseofaninterventionisanintegralpartofanevaluation.It is also important to ascertain the total number of users of a programme/intervention because it
providesan indicationof theneed fora service,and to findout thenumberofdropoutsbecause thisestablishestheextentoftherejectionoftheserviceforanyreason.Thereareanumberofproceduresforevaluatingtheparticipationofatargetpopulationinanintervention:
Percentageofusers–Theacceptanceofaprogrammebythetargetpopulationisoneoftheimportantindicatorsofaneedforit:thehighertheacceptance,thegreatertheneedfortheintervention.Somejudgethedesirabilityofaprogrammebythenumberofusersalone.Hence,asanevaluator,youcanexaminethetotalnumberofusersand,ifpossible,calculatethisasapercentageofthetotaltargetpopulation.However,youshouldbecarefulusingthepercentageofusersinisolationasanindicatorofthepopularityofaprogramme.Peoplemaybeunhappyanddissatisfiedwithaservice,yetuseitsimplybecausethereisnootheroptionavailabletothem.Ifusedwithotherindicators,suchasconsumersatisfactionorinrelationtoevidenceoftheeffectivenessofaprogramme,itcanprovideabetterindicationofitsacceptance.Percentageofeligibleusersofaservice–Servicerecordsusuallycontaininformationonserviceusersthatmayincludedataontheireligibilityfortheservice.Ananalysisofthisinformationwillprovideyouwithapercentageofeligibleusersoftheservice:thehigherthepercentageofeligibleusers,themorepositivetheevaluation.Thatis,
Youcanalsoundertakeasurveyoftheconsumersofaserviceinordertoascertainthepercentageofeligibleusers.Percentageofdropouts–Thedropoutratefromaserviceisreflectiveofthesatisfactionlevelofconsumerswiththeprogramme.Ahigherrateindicateseitherinappropriateservicecontentorflawsinthewaytheserviceisbeingdelivered:itdoesnotestablishwhethertheproblemiswiththedeliverymannerortheinterventioncontent.However,thefigurewillprovideyouwithanoverallindicationofthelevelofsatisfactionofconsumerswiththeservice:thehigherthedropoutrate,thehigherthelevelofdissatisfaction,eitherwiththecontentsofaservice(itsrelevancetotheneedsofthepopulation)orthewayitisbeingdelivered.
*Acceptorsareever-usersofaservice.
Surveyoftheconsumersofaservice–Ifservicerecordsdonotincludedataregardingclienteligibilityforaservice,youcanundertakeasurveyofever-users/acceptorsoftheservicetoascertaintheireligibilityfortheservice.Fromtheever-userssurveyed,youcanalsodeterminethedropoutrateamongthem.Inaddition,youcanfindoutmanyotheraspectsoftheevaluation,such
asclientsatisfaction,problemsandissueswiththeservice,orhowtoimproveitsefficiencyandeffectiveness.Howwellyoudothissurveyisdependentuponyourknowledgeofresearchmethodologyandavailabilityofresources.Surveyofthetargetpopulation–Targetpopulationsurveys,inadditiontoprovidinginformationabouttheextentofappropriateuseofaservice,alsoprovidedataontheextentofacceptanceofaserviceamongthoseforwhomitwasdesigned.Theproportionofpeoplewhohaveacceptedaninterventioncanbecalculatedasfollows:
Surveyofdropouts–Dropoutsareanextremelyusefulsourceofinformationforidentifyingwaysofimprovinganintervention.Thesearethepeoplewhohavegonethroughanintervention,haveexperiencedbothpositivesandnegatives,andhavethendecidedtowithdraw.Talkingtothemcanprovideyouwiththeirfirst-handexperienceoftheprogramme.Theyarethepeoplewhocanprovideyouwithinformationonpossibleproblems,eitherwiththecontentofaninterventionorwiththewayithasbeendelivered.Theyarealsoanexcellentsourceofsuggestionsonhowtoimproveaservice.Asurvey,focusgroupdiscussionorin-depthinterviewscanprovidevaluableinformationaboutthestrengthsaswellasweaknessesofaprogramme.Issuesraisedbythemandsuggestionsmademaybecomethebasisforimprovingthatintervention.Surveyofnon-usersofaservice–Whereasagroupofdropoutscanprovideextremelyusefulinformationabouttheproblemswithanintervention,non-usersareimportantinunderstandingwhysome,forwhomtheprogrammewasdesigned,havenotacceptedit.Chooseanymethod,quantitativeorqualitative,tocollectinformationfromthem.Ofcourseitcouldbeaproblemtoidentifythenon-usersinacommunity.
Evaluatingservicedeliverymanner
Therearesituationswhenaprogrammemaynothaveachieved its intendedgoals. Insuchsituations,therearetwopossiblecauses:thecontentoftheinterventionandthewayitisbeingdelivered.Itistomakesure thatan intervention isbeingdeliveredeffectively thatyouundertakeprocessevaluation. Itinvolvesidentifyingproblemswiththewayaserviceisbeingdeliveredtoconsumersorfindingwaysofimproving the delivery system. Evaluating the deliverymanner of a programme is a very importantaspectofprocessevaluation.Thereareanumberofissuesindeliveringaservicethatmayimpactuponitsdeliverymannerandprocessevaluationconsidersthem.Someoftheissuesare:
thedeliverymannerperse;thecontentsoftheserviceanditsrelevancetotheneedsofconsumers;theadequacyandqualityoftrainingimpartedtoserviceproviderstoenablethemtoundertakevarioustasks;staffmorale,motivationandinterestintheprogramme,andwaysofenhancingthese;theexpectationsofconsumers;resourcesavailableandtheirmanagement;issuesrelatingtoaccesstoservicesbythetargetpopulation;waysoffurtherimprovingthedeliveryofaservice.
Aprocessevaluationaimsatstudyingsomeoralloftheseissues.Thereareanumberofstrategiesthat
are used in process evaluation. The purpose for which you are going to use the findings shoulddeterminewhetheryouadoptaquantitativeorqualitativeapproach.Considerationsthatdeterminetheuseofqualitativeorquantitativemethodsingeneralalsoapplyinevaluationstudies.Methodsthatcanbeusedinundertakingaprocessevaluationare:
Opinionofconsumers–Oneofthebestindicatorsofthequalityofaserviceishowtheconsumersofthatservicefeelaboutit.Theyarebestplacedtoidentifyproblemsinthedeliverymanner,topointoutitsstrengthsandweaknesses,andtotellyouhowtheservicecanbeimprovedtomeettheirneeds.Simplybygatheringtheexperiencesofconsumerswithrespecttoutilisationofaserviceyoucangainvaluableinformationaboutitsstrengthsandweaknesses.Consumersurveysgiveyouaninsightintowhattheconsumersofaservicelikeanddonotlikeaboutaservice.Inthepresentageofconsumerismitisimportanttotaketheiropinionsintoconsiderationwhendesigning,deliveringorchangingaservice.Ifyouwanttoadoptaqualitativeapproachtoevaluation,youcanusein-depthinterviewing,focusgroupdiscussionsand/ortargetpopulationforumsaswaysofcollectinginformationabouttheissuesmentionedabove.Ifyouprefertouseaquantitativeapproachyoucanundertakeasurvey,givingconsiderationtoalltheaspectsofquantitativeresearchmethodologyincludingsamplesizeanditsselection,andmethodsofdatacollection.Keepinmindthatqualitativemethodswillprovideyouwithadiversityofopinionsandissuesbutwillnottellyoutheextentofthatdiversity.Ifyouneedtodeterminetheextentoftheseissues,youshouldcombinequalitativeandquantitativeapproaches.Opinionsofserviceproviders–Equallyimportantinprocessevaluationstudiesaretheopinionsofthoseengagedinprovidingaservice.Serviceprovidersarefullyawareofthestrengthsandweaknessesofthewayinwhichaprogrammeisbeingdelivered.Theyarealsowellinformedaboutwhatcouldbedonetoimproveinadequacies.Asanevaluator,youwillfindinvaluableinformationfromserviceprovidersforimprovingtheefficiencyofaservice.Again,youcanusequalitativeorquantitativemethodsfordatacollectionandanalysis.Time-and-motionstudies–Time-and-motionstudies,bothquantitativeandqualitative,canprovideimportantinformationaboutthedeliveryprocessofaservice.Thedominanttechniqueinvolvesobservingtheusersofaserviceastheygothroughtheprocessofusingit.You,asanevaluator,passivelyobserveeachinteractionandthendrawinferencesaboutthestrengthsandweaknessesofservicedelivery.Inaqualitativeapproachtoevaluationyoumainlyuseobservationasamethodofdatacollection,whereasinaquantitativeapproachyoudevelopmorestructuredtoolsfordatacollection(evenforobservation)andsubjectthedatatoappropriatestatisticalanalysisinordertomakeinferences.Functionalanalysisstudies–Analysisofthefunctionsperformedbyserviceprovidersisanotherapproachpeopleuseinthesearchforincreasedefficiencyinservicedelivery.Anobserver,withexpertiseinprogrammecontentandtheprocessofdeliveringaservice,followsaclientass/hegoesthroughtheprocessofreceivingit.Theobserverkeepsnoteofalltheactivitiesundertakenbytheserviceprovider,withthetimespentoneachofthem.Suchobservationsbecomethebasisforjudgingthedesirabilityofanactivityaswellasthejustificationforthetimespentonit,whichthenbecomesthebasisofidentifying‘waste’intheprocess.Again,youcanusequalitativeorquantitativemethodsofdatacollection.Youcanadoptveryflexiblemethodsofdatacollectionorhighlystructuredones.Youshouldbeawarethatobservationscanbeverystructuredorunstructured.Theauthorwasinvolvedinafunctional
analysisstudywhichinvolvedtwo-minuteobservationsofactivitiesofhealthworkersinacommunityhealthprogramme.Panelofexperts–Anothermethodthatisusedtostudythedeliveryprocessofaserviceistoaskexpertsintheareaofthatservicetomakerecommendationsabouttheprocess.Theseexpertsmayusevariousmethods(quantitativeorqualitative)togatherinformation,andsupplementitwiththeirownknowledge.Theythensharetheirexperiencesandassessmentswitheachotherinordertocomeupwithrecommendations.
The use of multiplemethodsmay providemore detailed and possibly better information but woulddepend upon the resources at your disposal and the purpose of your evaluation. Your skills as anevaluator lie in selecting amethod (ormethods) that best suits the purpose of evaluationwithin thegivenresources.
Impact/outcomeevaluation
Impact or outcome evaluation is one of themost widely practised types of evaluation. It is used toassesswhat changes can be attributed to the introduction of a particular intervention, programme orpolicy.Itestablishescausalitybetweenaninterventionanditsimpact,andestimatesthemagnitudeofthischange(s).Itplaysacentralroleindecisionmakingbypractitioners,managers,administratorsandplannerswhowish todeterminewhetherornotan interventionhasachieved itsdesiredobjectives inorder to make an informed decision about its continuation, termination or alteration.Many fundingorganisations base their decisions about further funding for programmes on impact evaluations.Specifically,anoutcomeevaluationisforthepurposeof:
establishingcausallinksbetweeninterventioninputsandoutcomes;measuringthemagnitudeoftheseoutcomes;determiningwhetheraprogrammeorinterventionhasachieveditsintendedgoals;findingouttheunintendedeffects,ifany,ofanintervention;comparingtheimpactsofaninterventionwithanalternativeoneinordertochoosethemoreeffectiveofthetwo.
Asyouareaware,inanycause-and-effectrelationship,inadditiontothecausetherearemanyotherfactorsthatcanaffecttherelationship.(FordetailsseeChapter7.)Justtorefreshyourmemory:
Inrelationtoaprogrammeorintervention,thisis
Thistheoryofcausality isofparticularrelevancetoimpactassessmentstudies.Indeterminingtheimpactofanintervention,itisimportanttorealisethatthechangesproducedbyaninterventionmaynotbesolelybecauseoftheintervention.Sometimes,otherfactors(extraneousvariables)mayplayamoreimportant role than the intervention in bringing about changes in the dependent variable.When youevaluate the effectiveness of an intervention, without comparing it to that of a control group, yourfindings will include the effects of extraneous variables. If you want to separate out the respectivecontributionsofextraneousvariablesandtheintervention,youneedtouseacontrolstudydesign.Therearemanydesignsfromwhichyoucanchooseinconductinganimpactassessmentevaluation.
Impact assessment studies range fromdescriptiveones– inwhichyoudescribepeople’s experiencesandperceptionsoftheeffectivenessofanintervention–torandom–control–blindexperiments.Again,your choice of a particular design is dependent upon the purpose of the evaluation and resourcesavailable.Someofthecommonlyuseddesignsare:
After-onlydesign–Thoughtechnicallyinappropriate,after-onlydesignisacommonlyuseddesigninevaluationstudies.Itmeasurestheimpactofaprogrammeorintervention(afterithasoccurred)withouthavingabaseline.Theeffectivenessoftheinterventionisjudgedonthebasisofthecurrentpictureofthestateofevaluationindicators.Itreliesonindicatorssuchas:
numberofusersoftheservice;numberofdropoutsfromtheservice;satisfactionofclientswiththeservice;stories/experiencesofclientsthatchangedthem;assessmentmadebyexpertsinthearea;theopinionsofserviceproviders.
Itisonthebasisoffindingsabouttheseoutcomeindicatorsthatadecisionaboutcontinuation,terminationoralterationsinaninterventionismade.Oneofthemajordrawbacksofthisdesignisthatitdoesnotmeasurechangethatcanbeattributedtotheinterventionassuch,since(asmentioned)ithasneitherabaselinenoracontrolgrouptocompareresultswith.However,itprovidesthecurrentpictureinrelationtotheoutcomeindicators.Thisdesignisthereforeinappropriatewhenyouareinterestedinstudyingtheimpactofaninterventionperse.Before-and-afterdesign–Thebefore-and-afterdesignistechnicallysoundandappropriateformeasuringtheimpactofanintervention.Therearetwowaysofestablishingthebaseline.Onewayiswherethebaselineisdeterminedbeforetheintroductionofanintervention,whichrequiresadvanceplanning;andtheotheriswherethebaselineisestablishedretrospectively,eitherfrompreviousservicerecordsorthroughrecallbyclientsoftheirsituationbeforetheintroductionoftheintervention.Retrospectiveconstructionofthebaselinemayproducelessaccuratedatathanafterthedatacollectionandhencemaynotbecomparable.However,intheabsenceofanythingbetter,itdoesprovidesomebasisofcomparison.Asyoumayrecall,oneofthedrawbacksofthisdesignisthatthechangemeasuredincludeschangebroughtaboutbyextraneousandchangevariables.Hence,thisdesign,thoughacceptableandbetterthantheafter-onlydesign,stillhasatechnicalproblemintermsofevaluationstudies.Also,itismoreexpensivethantheafter-onlydesign.Experimental–controldesign–Thebefore-and-afterstudy,withacontrolgroup,isprobablytheclosesttoatechnicallycorrectdesignforimpactassessmentofanintervention.Oneofthebiggest
strengthsofthisdesignisthatitenablesyoutoisolatetheimpactofindependentandextraneousvariables.However,itaddstheproblemofcomparabilitybetweencontrolandexperimentalgroups.Sometimesthisproblemofcomparabilitycanbeovercomebyformingthegroupsthroughrandomisation.Unfortunately,complexityinitsexecutionandincreasedcostrestricttheuseofthisdesignfortheaverageevaluationstudy.Also,inmanysituationsitmaynotbepossibletofindorconstructasuitablecontrolgroup.Comparativestudydesign–Thecomparativestudydesignisusedwhenevaluatingtwoormoreinterventions.Forcomparativestudiesyoucanfollowanyoftheabovedesigns;thatis,youcanhaveacomparativestudyusingafter-only,before-and-afterorexperimental–controldesign.Reflexivecontroldesign–Toovercometheproblemofcomparabilityindifferentgroups,sometimesresearcherstreatdatacollectedduringthenon-interventionperiodtorepresentacontrolgroup,andinformationcollectedaftertheintroductionoftheinterventionasifitpertainedtoanexperimentalgroup(Figure18.5).Inthereflexivecontroldesign,comparisonbetweendatacollection2anddatacollection1providesinformationforthecontrolgroup,whilecomparisonbetweendatacollection3anddatacollection2providesdatafortheexperimentalgroup.Oneofthemainadvantagesofthisdesignisthatyoudonotneedtoensurethecomparabilityoftwogroups.However,iftherearerapidchangesinthestudypopulationovertime,andiftheoutcomevariablesarelikelytobeaffectedsignificantly,useofthisdesigncouldbeproblematic.
FIGURE18.5ReflexivecontroldesignInterruptedtime-seriesdesign–Intheinterruptedtime-seriesdesignyoustudyagroupofpeoplebeforeandaftertheintroductionofanintervention.Itislikethebefore-and-afterdesign,exceptthatyouhavemultipledatacollectionsatdifferenttimeintervalstoconstituteanaggregatedbefore-and-afterpicture(Figure18.6).Thedesignisbasedupontheassumptionthatonesetofdataisnotsufficienttoestablish,withareasonabledegreeofcertaintyandaccuracy,thebefore-and-aftersituations.
FIGURE18.6Interruptedtime-seriesdesign
Replicatedcross-sectionaldesign–Thereplicatedcross-sectionaldesignstudiesclientsatdifferentstagesofanintervention,andisappropriateforthoseinterventionsthattakenewclientsonacontinuousorperiodicbasis.SeeFigure18.7.Thisdesignisbasedontheassumptionthatthosewhoarecurrentlyattheterminationstageofaninterventionaresimilarintermsofthenatureandextentoftheproblemtothosewhoarecurrentlyattheintakestage.Inordertoascertainthechangethatcanbeattributedtoanintervention,asampleattheintakeandterminationstagesoftheprogrammeisselected,sothatinformationcanbecollectedpertainingtopre-situationsandpost-situationswithrespecttotheproblemforwhichtheinterventionisbeingsought.Toevaluatethepatternofimpact,sometimesresearcherscollectdataatoneormoreintermediarystages.
These designs vary in sophistication and so do the evaluation instruments. Choice of design isdifficultand(asmentionedearlier)itdependsuponthepurposeandresourcesavailable.
FIGURE18.7Replicatedcross-sectionaldesign
Anotherdifficultyistodecidewhen,duringtheinterventionprocess,toundertaketheevaluation.Howdo you know that the intervention has made its impact? One major difficulty in evaluating socialprogrammesrevolvesaroundthequestion:wasthechangeaproductoftheinterventionordiditcomefromaconsumer’srelationshipwithaserviceprovider?Manysocialprogrammesareacceptedbecauseof theconfidenceconsumersdevelop inaserviceprovider. Inevaluationstudiesyouneed tokeep inmindtheimportanceofaserviceproviderinbringingaboutchangeinindividuals.
Cost–benefit/cost-effectivenessevaluation
Whileknowledgeabouttheprocessofservicedeliveryanditsoutcomesishighlyusefulforanefficientand effective programme, in some cases it is critical to be informed about how intervention costscomparewith outcomes. In today’sworld,which is characterised by scarce resources and economicrationalism,itisimportanttojustifyaprogrammeinrelationtoitscost.Cost–benefitanalysisprovidesa framework for relating costs to benefits in terms of a common unit ofmeasurement,monetary orotherwise.Specifically, cost–benefit analysisorcost-effectivenessevaluation is important because ithelpsto:
putlimitedresourcestooptimaluse;decidewhichoftwoequallyeffectiveinterventionstoreplicateonalargerscale.
Cost–benefitanalysisfollowsaninput/outputmodel,thequalityofwhichdependsupontheabilitytoidentify accurately and measure all intervention inputs and outputs. Compared with technicalinterventions,suchasthosewithinengineering,socialinterventionsaremoredifficulttosubjecttocost–
benefit analysis.This is primarily because of the difficulties in accurately identifying andmeasuringinputs and outputs, and then converting them to a commonmonetary unit. Some of the problems inapplyingcost–benefitanalysistosocialprogrammesareoutlinedbelow:
Whatconstitutesaninputforanintervention?Therearedirectandindirectinputs.Identifyingthesecansometimesbeverydifficult.Evenifyouhavebeenabletoidentifythem,thenextproblemisputtingapricetagoneachofthem.Similarly,theoutputsorbenefitsofaninterventionneedtobeidentifiedandmeasured.Likeinputs,benefitscanalsobedirectandindirect.Inaddition,aprogrammemayhaveshort-termaswellaslong-termbenefits.Howdoyoucostthevariousbenefitsofaprogramme?Anothercomplexityistheneedtoconsiderbenefitsfromtheperspectivesofdifferentstakeholders.Themainproblemincost–benefitanalysisisthedifficultyinconvertinginputsaswellasoutputstoacommonunit.Insocialprogrammes,itoftenbecomesdifficulteventoidentifyoutputs,letalonemeasureandthenconvertthemtoacommonunitofmeasurement.
Typesofevaluationfromaphilosophicalperspective
Fromaphilosophicalperspective, therearenospecificmodelsforormethodsofevaluation.Youusethesamemethodsandmodelsbuttherequiredinformationisgatheredfromdifferentpeopleoraspectsdepending upon the philosophy that you subscribe to. Stufflebeam and Shinkfield’s bookSystematicEvaluation:ASelf-InstructionalGuide toTheoryandPractice is anexcellent source toacquaintyouwiththeseperspectives.Typesofevaluationcategorisedonthebasisofphilosophies,mentionedbelow,aredealtwithingreaterdetailintheirbookanditishighlyrecommendedthatyourefertothatifyouwant togainabetter appreciationof theseperspectives.On thebasisof theseperspectives, therearefourtypesofevaluation.Again,youshouldkeepinmindthatthisclassificationandtheclassificationdevelopedonthebasisofthefocusofevaluationarenotmutuallyexclusive.
Goal-centred/objective-orientedevaluation
This approach is based upon the philosophy that the success or failure of an intervention should bebasedupontheextentofcongruencebetweentheobjectivesofaninterventionanditsactualoutcomes.Thisapproachstudiesoutcomes todetermine theachievementofobjectives,andcongruencebetweenthe two is regarded as the sole determinant of success or failure. One of the main criticisms ofobjective-orientedevaluation is that itassessestheeffectivenessofaprogrammewithoutexplainingthereasonsforit.Basically, the process of evaluation involves, firstly, identification of the desired goals of an
interventionand,secondly,theuseofaprocesstomeasuretheirsuccessorfailure.Again,youcanuseeitherqualitativeorquantitativemethodstoachievethis.
Consumer-oriented/client-centredevaluation
The core of this philosophy rests on the assumption that assessment of the value or merit of anintervention–includingitseffectiveness,outcomes,impactandrelevance–shouldbejudgedfromtheperspective of the consumer. Consumers, according to the philosophy of consumer-oriented
evaluation,arethebestjudgesofaprogramme.Client-centredevaluations,again,mayusequalitativeorquantitativemethodstofindouthowclients
feelaboutvariousaspectsofanintervention.Youcanevenuseamixofthetwotofindoutconsumers’perceptionsandopinions.
Improvement-orientedevaluation
The basic philosophy behind improvement-oriented evaluation is that an evaluation should fosterimprovement. ‘Not to prove but to improve’ seems to be the central themeof such evaluations.Thefocusistostudythecontextinordertohelpimproveaninterventioncontent–theprocessratherthanoutcomes.Again,amultiplicityofmethodscanbeusedtoundertakesuchevaluation.
Holistic/illuminativeevaluation
Theprimaryconcernofholisticresearchorilluminativeevaluationisdescriptionandinterpretation,rather than measurement and prediction. It fits with the social–anthropological paradigm,acknowledgingasitdoeshistorical,culturalandsocialfactorswhenevaluatinganintervention.Theaimistostudyaprogrammeinallitsaspects:howitoperates,howitisinfluencedbyvariouscontexts,howitisapplied,howthosedirectlyinvolvedviewitsstrengthsandweaknesses,andwhattheexperiencesare of thosewho are affected by it. In summary, it tries to illuminate a complex array of questions,issues and factors, and to identify procedures that give both desirable and undesirable results. So aholistic/illuminative evaluation tries to understand issues relating to an intervention from manyperspectives:itseekstoviewtheperformanceofaprogrammeinitstotality.Anevaluationcanbeconductedfromanyoneoftheabovephilosophicalperspectives.Tous,these
areperspectivesrather thanevaluationmodels,butsomeusethemastypesofevaluation.Theaimofthissectionhasbeentoacquaintyouwithsomeoftheseperspectives.
Undertakinganevaluation:theprocess
Liketheresearchmethodologymodel,whichformsthebasisofthisbook,theevaluationprocessisalsobaseduponcertainoperationalsteps.Itisimportantforyoutorememberthattheorderinthewrite-upof thesesteps isprimarily tomake iteasier foryou tounderstand theprocess.Onceyouare familiarwiththesesteps,theirordercanbechanged.
Step1:Determiningthepurposeofevaluation
In a research study you formulate your research problem before developing a methodology. In anevaluation study too, youneed to identify thepurposeof undertaking it anddevelopyourobjectivesbeforeventuringintoit.Itisimportanttoseekanswerstoquestionssuchas:‘WhydoIwanttodothisevaluation?’ and ‘Forwhatpurposewould I use the findings?’Specifically, youneed to consider thefollowingmatters, and to identify their relevance and application to your situation. Is the evaluationbeingundertakentodothefollowing?
Identifyandsolveproblemsinthedeliveryprocessofaservice.Increaseefficiencyoftheservicedeliverymanner.Determinetheimpactsoftheintervention.Trainstaffforbetterperformance.Workoutanoptimalworkloadforstaff.Findoutaboutclientsatisfactionwiththeservice.Seekfurtherfunding.Justifycontinuationoftheprogramme.Resolveissuessoastoimprovethequalityoftheservice.Testoutdifferentinterventionstrategies.Choosebetweentheinterventions.Estimatethecostofprovidingtheservice.
Itisimportantthatyouidentifythepurposeofyourevaluationandfindanswerstoyourreasonsforundertaking it with the active involvement and participation of the various stakeholders. It is alsoimportantthatallstakeholders–clients,serviceproviders,servicemanagers,fundingorganisationsandyou,asanevaluator–agreewiththeaimsoftheevaluation.Makesurethatallstakeholdersalsoagreethatthefindingsoftheevaluationwillnotbeusedforanypurposeotherthanthoseagreedupon.Thisagreement is important in ensuring that the findings will be acceptable to all, and for developingconfidenceamongthosewhoaretoprovidetherequiredinformationdosofreely.Ifyourrespondentsarescepticalabouttheevaluation,youwillnotobtainreliableinformationfromthem.Havingdecidedonthepurposeofyourevaluation,thenextstepistodevelopasetofobjectivesthat
willguideit.
Step2:Developingobjectivesorevaluationquestions
Asinaresearchproject,youneedtodevelopevaluationquestions,whichwillbecomethefoundationfortheevaluation.Well-articulatedobjectivesbringclarityandfocustothewholeevaluationprocess.Theyalsoreducethechancesofdisagreementlateramongvariousparties.Someorganisationsmaysimplyaskyou‘toevaluatetheprogramme’,whereasothersmaybemuch
morespecific.Thesamemaybethesituationifyouareinvolvedinevaluatingyourownintervention.Ifyou have been given specific objectives or you are in a situation where you are clear about theobjectives,youdonotneedtogothroughthisstep.However,ifthebriefisbroad,oryouarenotclearabouttheobjectivesinyourownsituation,youneedtoconstructforyourselfandothersa‘meaning’ofevaluation.As you know, evaluation can mean different things to different people. To serve the purpose of
evaluationfromtheperspectivesofdifferentstakeholders,itisimportanttoinvolveallstakeholdersinthedevelopmentofevaluationobjectivesandtoseektheiragreementwiththem.Youneedtofollowthesame process as for a research study (Chapter 4). The examples in Figure 18.8 may help you tounderstandmoreaboutobjectiveformulation.
Example:Developingevaluationobjectives:examplesRecently the author was asked to undertake two evaluations. For one, the brief was ‘To evaluate the principle of communityresponsiveness in thedeliveryofhealth in…(nameof thestate)’,andfor theother itwas‘Toevaluate…(nameof themodel)servicedeliverymodelin…(nameoftheregion)’.
Evaluatingaprogramme:ExampleOneFor the first evaluation, after having initial discussions with various stakeholders, it was discovered that understanding of theprinciple of ‘community responsiveness’was extremely vague and varied among different people.Also, therewere neither anyinstructions about how to achieve community responsiveness nor any training programme for the purpose. A few people,responsibleforensuringtheimplementationoftheprinciple,hadnoideaaboutitsimplementation.Ourfirstquestionwas:‘Canweevaluatesomethingaboutwhichthoseresponsibleforimplementationarenotclear,andforwhichthereisnospecificstrategyinplace?’Theobvious answerwas ‘no’.Wediscussedwith the sponsorsof the evaluationwhat questions theyhad inmindwhenaskingus for theevaluation.On thebasisofourdiscussionwith themandourunderstandingof their reasons for requesting theevaluation,weproposedthattheevaluationbecarriedoutintwophases.Forthefirstphase,theaimoftheevaluationshouldbetodefine‘communityresponsiveness’, identify/develop/exploreoperationalstrategies toachieve it,and identify the indicatorsof itssuccess or otherwise. During the second phase, an evaluation to measure the impact of implementation of the communityresponsivenessstrategieswasproposed.Ourproposalwasaccepted.Wedevelopedthefollowingobjectivesinconsultationwiththevariousstakeholders.
EvaluationoftheprincipleofcommunityresponsivenessinhealthPhaseOne
Mainobjective:Todevelopamodelforimplementingtheprincipleofcommunityresponsivenessinthedeliveryofhealthcarein…(nameofthestate).Specificobjectives:
1. Tofindouthowtheprincipleofcommunityresponsivenessisunderstoodbyhealthplanners,administrators,managers,serviceprovidersandconsumers,andtodevelopanoperationaldefinitionofthetermforthedepartment.
2. Toidentify,withtheparticipationofstakeholders,strategiestoimplementtheconceptofcommunityresponsivenessinthedeliveryofhealthservices.
3. Todevelopasetofindicatorstoevaluatetheeffectivenessofthestrategiesusedtoachievecommunityresponsiveness.
4. Toidentifyappropriatemethodologiesthatareacceptabletostakeholdersformeasuringeffectivenessindicators.
PhaseTwo
Mainobjective:Toevaluatetheeffectivenessofthestrategiesusedtoachievetheprincipleofcommunityresponsivenessinthedeliveryofhealthservices.
Subobjectives:
1. Todeterminetheimpactofcommunityresponsivenessstrategiesoncommunityparticipationindecisionmakingabouthealthissuesaffectingthecommunity.
2. Tofindouttheopinionsofthevariousstakeholdersonthedegreetowhichtheprovisionofcommunityresponsivenessinthedeliveryofhealthserviceshasbeen/isbeingobserved.
3. Tofindouttheextentofinvolvementofthecommunityindecisionmakinginissuesconcerningthecommunityanditsattitudetowardsinvolvement.
Evaluatingaprogramme:ExampleTwoNowletustakethesecondstudy.Inthiscasetheservicedeliverymodelwaswelldevelopedandtheevaluationbriefwasclearinterms of its expectations; that is, the objective was to evaluate the model’s effectiveness. Before starting the evaluation, thefollowingobjectivesweredeveloped inconsultationwith thesteeringcommittee,whichhad representatives fromall stakeholdergroups. Remember, it is important thatyourobjectivesbeunambiguous, clear and specific, and that theyarewrittenusingverbs thatexpressyouroperationalintentions.
The…ModelMainobjective:Toevaluatetheeffectivenessofthe…(nameofthemodel)developedby…(nameoftheoffice).
Subobjectives:
1. Toidentifythestrengthsandweaknessesofthemodelasperceivedbyvariousstakeholders.2. Tofindouttheattitudesofconsumers,serviceprovidersandmanagers,andrelevant
communityagenciestowardsthemodel.3. Todeterminetheextentofreduction,ifany,inthenumberofchildreninthecareofthe
departmentsincetheintroductionofthemodel.4. TodeterminetheimpactofthemodelonthenumberofChildConcernReportsandChild
MaltreatmentAllegations.5. Toassesstheabilityofthemodeltobuildthecapacityofconsumersandserviceprovidersto
dealwithproblemsintheareaofchildprotection.6. Torecommendstrategiestoovercomeproblems,ifany,withthemodel.7. Toestimatethecostofdeliveringservicesinaccordancewiththemodeltoafamily.
Step3:Convertingconceptsintoindicatorsintovariables
Inevaluation,aswellasinotherresearchstudies,oftenweuseconceptstodescribeourintentions.Forexample,we say thatweare seeking to evaluateoutcomes, effectiveness, impactor satisfaction.Themeaningascribedtosuchwordsmaybecleartoyoubutmaydiffermarkedlyfromtheunderstandingofothers.Thisisbecausethesetermsinvolvesubjectiveimpressions.Theyneedoperationaldefinitionsintermsoftheirmeasurementinordertodevelopauniformunderstanding.Whenyouuseconcepts,thenextproblemyouneedtodealwithisthedevelopmentofa‘meaning’foreachconceptthatdescribesthem appropriately for the contexts inwhich they are being applied.Themeaning of a concept in aspecific situation is arrived at by developing indicators. To develop indicators, you must answerquestionssuchas:‘Whatdoesthisconceptmean?’,‘WhencanIsaythattheprogrammeiseffective,orhasbroughtaboutachange,orconsumersorserviceprovidersaresatisfied?’and‘OnwhatbasisshouldIconcludethataninterventionhasbeeneffective?’Answerstosuchquestionsbecomeyourindicatorsand theirmeasurementandassessmentbecome thebasisof judgementabouteffectiveness, impactorsatisfaction. Indicators are specific, observable, measurable characteristics or changes that can beattributedtotheprogrammeorintervention.A critical challenge to an evaluator in outcome measurement is identifying and deciding what
indicators to use in order to assess howwell the programme being evaluated has done regarding anoutcome.Rememberthatnotallchangesorimpactsofaprogrammemaybereflectedbyoneindicator.Inmanysituationsyouneedtohavemultipleindicatorstomakeanassessmentofthesuccessorfailureofaprogramme.Figure18.9showstheprocessofconvertingconcepts intoquestions thatyouaskofyourrespondents.Someindicatorsareeasytomeasure,whereasothersmaybedifficult.Forexample,anindicatorsuch
asthenumberofprogrammeusersiseasytomeasure,whereasaprogramme’simpactonself-esteemismoredifficulttomeasure.In order to assess the impact of an intervention, different types of effectiveness indicators can be
used.Theseindicatorsmaybeeitherqualitativeorquantitative,andtheirmeasurementmayrangefrom
subjective–descriptiveimpressionstoobjective–measurable–discretechanges.Ifyouareinclinedmoretowardsqualitativestudies,youmayusein-depthinterviewing,observationorfocusgroupstoestablishwhetherornottherehavebeenchangesinperceptions,attitudesorbehaviouramongtherecipientsofaprogrammewithrespecttotheseindicators.Inthiscase,changesareasperceivedbyyourrespondents:there is,assuch,nomeasurement involved.On theotherhand, ifyoupreferaquantitativeapproach,youmayusevariousmethodstomeasurechangeintheindicatorsusingintervalorratioscales.Inallthedesignsthatwehavediscussedaboveinoutcomeevaluation,youmayusequalitativeorquantitativeindicatorstomeasureoutcomes.
FIGURE18.8ConvertingconceptsintoindicatorsintovariablesNowletustakeanexampletoillustratetheprocessofconvertingconceptstoquestions.Supposeyou
are working in a department concerned with protection of children and are testing a newmodel ofservice delivery. Let us further assume that your model is to achieve greater participation andinvolvement of children, their families and non-statutory organisationsworking in the community indecisionmaking about children.Your assumption is thatwith their involvement and participation indevelopingtheproposedinterventionstrategies,highercompliancewillresult,which,inturn,willresultintheachievementofthedesiredgoals.Aspartofyourevaluationofthemodel,youmaychooseanumberofindicatorssuchastheimpact
onthe:
numberofchildrenunderthecareofthedepartment/agency;numberofchildrenreturnedtothefamilyorthecommunityforcare;numberofreportedcasesof‘ChildMaltreatmentAllegations’;numberofreportedcasesof‘ChildConcernReports’;extentofinvolvementofthefamilyandcommunityagenciesinthedecision-makingprocessaboutachild.
Youmayalsochooseindicatorssuchastheattitudeof:
children,whereappropriate,andfamilymemberstowardstheirinvolvementinthedecision-makingprocess;serviceprovidersandservicemanagerstowardstheusefulnessofthemodel;non-statutoryorganisationstowardstheirparticipationinthedecision-makingprocess;variousstakeholderstowardstheabilityofthemodeltobuildthecapacityofconsumersoftheserviceforself-management;familymemberstowardstheirinvolvementinthedecision-makingprocess.
Thescalesused in themeasurementdeterminewhetheran indicatorwillbeconsideredas ‘soft’or‘hard’.Attitudetowardsanissuecanbemeasuredusingwell-advancedattitudinalscalesorbysimplyasking a respondent to give his/her opinion. The first method will yield a hard indicator while the
secondwillprovidea softone.Similarly,achange in thenumberofchildren, ifaskedasanopinionquestion,willbetreatedasasoftindicator.
FIGURE18.9AnexampleofconvertingconceptsintoquestionsFigure 18.10 summarises the process of converting concepts into questions, using the example
described above.Once you have understood the logic behind this operationalisation, youwill find iteasiertoapplyinothersimilarsituations.
Step4:Developingevaluationmethodology
Aswithanon-evaluative study,youneed to identify thedesign thatbest suits theobjectivesofyourevaluation,keepinginmindtheresourcesatyourdisposal.Inmostevaluationstudiestheemphasisison‘constructing’acomparativepicture,beforeandaftertheintroductionofanintervention,inrelationtotheindicatorsyouhaveselected.Onthebasisofyourknowledgeaboutstudydesignsandthedesignsdiscussed in this chapter, you propose one that is most suitable for your situation. Also, as part ofevaluationmethodology,donotforgettoconsiderotheraspectsoftheprocesssuchas:
Fromwhomwillyoucollecttherequiredinformation?Howwillyouidentifyyourrespondents?Areyougoingtoselectasampleofrespondents?Ifyes,howandhowlargewillitbe?Howwillyoumakeinitialcontactwithyourpotentialrespondents?Howwillyouseektheinformedconsentofyourrespondentsfortheirparticipationinthe
evaluation?Howwilltheneededinformationbecollected?Howwillyoutakecareoftheethicalissuesconfrontingyourevaluation?Howwillyoumaintaintheanonymityoftheinformationobtained?Whatistherelevanceoftheevaluationforyourrespondentsorothersinasimilarsituation?
Youneedtoconsideralltheseaspectsbeforeyoustartcollectingdata.
Step5:Collectingdata
Asinaresearchstudy,datacollectionisthemostimportantandtime-consumingphase.Asyouknow,thequalityofevaluationfindingsisentirelydependentuponthedatacollected.Hence,theimportanceofdatacollectioncannotbeoveremphasised.Whetherquantitativeorqualitativemethodsareusedfordatacollection,itisessentialtoensurethatqualityismaintainedintheprocess.You can have a highly structured evaluation, placing great emphasis on indicators and their
measurement,oryoucanoptforanunstructuredandflexibleenquiry:asmentionedearlier,thedecisionisdependentupon thepurposeofyourevaluation.Forexploratorypurposes, flexibilityanda lackofstructure are an asset, whereas, if the purpose is to formulate a policy, measure the impact of aninterventionortoworkoutthecostofanintervention,agreaterstructureandstandardisationandlessflexibilityareimportant.
Step6:Analysingdata
Aswithresearchingeneral, thewayyoucananalysethedatadependsuponthewayitwascollectedandthepurposeforwhichyouaregoingtousethefindings.Forpolicydecisionsanddecisionsaboutprogramme termination or continuation, you need to ascertain themagnitude of change, based on areasonablesamplesize.Hence,yourdataneedstobesubjectedtoastatisticalframeworkofanalysis.However,ifyouareevaluatingaprocessorprocedure,youcanuseaninterpretiveframeofanalysis.
Step7:Writinganevaluationreport
Aspreviouslystated, thequalityofyourworkand the impactofyour findingsaregreatlydependentuponhowwellyoucommunicatethemtoyourreaders.Yourreportistheonlybasisofjudgementforanaveragereader.Hence,youneedtopayextraattentiontoyourwriting.As for a research report, there are different writing styles. In the author’s opinion you should
communicate your findings under headings that reflect the objectives of your evaluation. It is alsosuggestedthatthefindingsbeaccompaniedbyrecommendationspertainingtothem.Yourreportshouldalsohaveanexecutivesummaryofyourfindingsandrecommendations.
Step8:Sharingfindingswithstakeholders
A very important aspect of any evaluation is sharing the findings with the various groups ofstakeholders. It is a good idea to convene a group comprising all stakeholders to communicatewhatyour evaluation has found.Be open about your findings and resist pressure from any interest group.Objectivelyandhonestlycommunicatewhatyourevaluationhasfound.Itisofutmostimportancethat
youadheretoethicalprinciplesandtheprofessionalcodeofconduct.Asyouhaveseen,theprocessofaresearchstudyandthatofanevaluationisalmostthesame.The
onlydifferenceistheuseofcertainmodelsinthemeasurementoftheeffectivenessofanintervention.Itisthereforeimportantforyoutoknowaboutresearchmethodologybeforeundertakinganevaluation.
Involvingstakeholdersinevaluation
Most evaluations have a number of stakeholders, ranging from consumers to experts in the area,includingserviceprovidersandmanagers.Itisimportantthatallcategoriesofstakeholderbeinvolvedat all stages of an evaluation. Failure to involve any groupmayhinder success in completion of theevaluationandseriouslyaffectconfidenceinyourfindings.Itisthereforeimportantthatyouidentifyallstakeholdersandseektheirinvolvementandparticipationintheevaluation.Thisensuresthattheyfeelapartoftheevaluationprocess,which,inturn,markedlyenhancestheprobabilityoftheiracceptingthefindings.Thefollowingstepsoutlineaprocessforinvolvingstakeholdersinanevaluationstudy.
Step1
Identifyingstakeholders.Firstofall,talkwithmanagers,planners,programmeadministrators,serviceprovidersandtheconsumersoftheprogrammeeitherindividuallyorcollectively,andidentifywhotheythinkarethedirectandindirectstakeholders.Havingcollectedthisinformation,shareitwithallgroupsofstakeholderstoseeifanyonehasbeenleftout.Preparealistofallstakeholdersmakingsureitisacceptabletoallsignificantones.Ifthereareanydisagreements,itisimportanttoresolvethem.
Step2
Involvingstakeholders.Inordertodevelopacommonperspectivewithrespecttovariousaspectsoftheevaluation,itisimportantthat different categories of stakeholder be actively involved in thewhole process of evaluation from the identification of theirconcerns to the sharingof its findings. Inparticular, it is important to involve them indeveloping a framework for evaluation,selectingtheevaluationindicators,anddevelopingproceduresandtoolsfortheirmeasurement.
Step3
Developing a common perspective among stakeholders towards the evaluation. Different stakeholders may have differentunderstandingsoftheword‘evaluation’.Somemayhaveaverydefiniteopinionaboutitandhowitshouldbecarriedoutwhileothersmaynothaveanyconception.Differentstakeholdersmayalsohavedifferentopinionsabout therelevanceofaparticularpieceofinformationforansweringanevaluationquestion.Ortheymayhavedifferentinterests.Tomakeevaluationmeaningfultothemajority of stakeholders, it is important that their perspectives and understandings of evaluation be understood and that acommonperspectiveontheevaluationbearrivedatduringtheplanningstage.
Step4
Resolvingconflictsofinterest.Asanevaluator,ifyoufindthatstakeholdershavestrongopinionsandthereisaconflictofinterestamong themwith respect toanyaspectof theevaluation, it isextremely important to resolve it.However,youhave tobeverycarefulinresolvingdifferencesandmustnotgivetheimpressionthatyouarefavouringanyparticularsubgroup.
Step5
Identifyingthe informationstakeholdersneedfromtheproposedevaluation. Identify, fromeachgroupofstakeholders, theinformationtheythinkisimportanttomeettheirneedsandtheobjectivesoftheevaluation.
Step6
Formingasteeringcommittee.Forroutineconsultation,thesharingofideasandday-to-daydecisionmaking,itisimportantthatyouask thestakeholders toelectasteeringcommitteewithwhomyou,as theevaluator,canconsultandinteract. Inaddition toprovidingyouwithaforumforconsultationandguidance,suchacommitteegivesstakeholdersacontinuoussenseofinvolvementintheevaluation.
Ethicsinevaluation
Beingethicalisthecorerequirementofanevaluation.Ifforsomereasonyoucannotbeethical,donotundertaketheevaluation,asyouwillendupdoingharmtoothers,andthatisunethical.Although,asagood evaluator, you may have involved all the stakeholders in the planning and conduct of theevaluation,itispossiblethatsometimes,whenfindingsarenotinsomeone’sinterest,astakeholderwillchallengeyou.Itisoftheutmostimportancethatyoustandfirmonthefindingsanddonotsurrenderto
anypressurefromanyone.Surrenderingtosuchpressureisunethical.
SummaryInthischaptersomeoftheaspectsofevaluationresearcharediscussed,inbrief,inordertomakeyouawareofthem,ratherthantoprovideyouwithadetailedknowledgebase.Itishighlyrecommendedthatyoureadsomebooksonevaluationresearch.Thischapterhighlights the relationshipbetween researchmethodologyper seand itsapplication toevaluation inpractice.Evaluationskillsarebuiltontheknowledgeandskillsofresearchmethodology:anevaluatorhastobeagoodresearcher.In this chapterwe looked at some of the definitions of ‘evaluation’, identified its characteristics and examined the reasons for
undertaking an evaluation. The intervention–development–evaluation process is discussed in detail, exploring the relationshipbetweenprogrammedevelopmentanditsevaluation.Evaluationstudiesareclassifiedfromtwoperspectives:thefocusofevaluationandthephilosophicalbasisthatunderpinsthem.Thetypologyofevaluationstudiesisdevelopedfromtheseperspectives.Therearefour different types of evaluation from the perspective of their focus: programme/intervention planning evaluation,process/monitoringevaluation,impact/outcomeevaluationandcost–benefit/cost-effectivenessevaluation.Fromtheperspectiveofthephilosophies that underpin these evaluations, again, four types of evaluation are identified: goal-centred/objective evaluation,consumer-oriented/client-centredevaluation,improvement-oriented evaluation andholistic evaluation. The evaluation processwasoutlined stepby stepwith considerablediscussion centredonhow to convert concepts into indicators intovariables, enabling theformulation of questions for respondents that will elicit the required information. How to involve stakeholders in an evaluationprocesswasalsodiscussedusingastep-by-stepguide.Finally,thereadersarealertedtosomeoftheethicalissuesinevaluation.
ForYoutoThinkAbout
Refamiliariseyourselfwiththekeywordslistedatthebeginningofthischapterandifyouareuncertainaboutthemeaningorapplicationofanyofthemrevisittheseinthechapterbeforemovingon.Imaginethatyouhavebeenaskedtoevaluateaserviceofferedbytheorganisationyouworkfor.Considerhowyouwouldgoaboutthisprocesstakingintoaccountanyethicaldilemmasthatmayariseandthepracticalproblemsthatyoumayface.Takinganexampleofanevaluationstudyfromyourownareaofinterestorprofession,identifythestakeholdersandconsiderwhyitisimportanttoinvolvethemintheprocess.Why,asaserviceprovider,isitimportantthatyouevaluateyourownpractice?
AppendixDevelopingaresearchproject:asetofexercisesforbeginners
Application is the essence of knowledge. However, there always remains a gap between theoreticalknowledge and its application. It is only with practice that this gap can be narrowed. A beginnerattemptingtoapplytheoreticalknowledgeneedsdirectionandguidance.Thissetofexerciseshasbeendeveloped with this belief. There is an exercise for almost each operational step of the proposedresearchprocess.Workingthroughthemwillhelpyoutodeveloparesearchproject.Themain aim of these exercises is to provide youwith a broad framework that is central to the
operationalisationofeachstepoftheresearchprocess.Inmostcases,aseparateexerciseisprovidedforquantitativeandqualitativestudiessoitisimportantthatyouknowbeforeyoustartwhichapproachyouaregoingtotake.Withineachexercise,therearebriefremindersofsomeofthekeyissuesrelatingtotheprocessandaseriesofquestionstohelpyoutothinkthroughproceduresandprovideaframeworkforthedevelopmentofyourstudy.Answers to thesequestionsandawarenessof the issues that theexercisesoutlinewillputyou ina
position to complete the framework suggested for writing a research proposal (Chapter 13), andthereforethesewillalsoconstitutethecoreofyourresearchproposal.Itisimportantforabeginnertoworkthroughtheseexerciseswithconsiderablethoughtandcare.
ExerciseI:Formulationofaresearchproblem
Quantitativestudies
NowthatyouhavegonethroughallthechaptersthatconstituteStepIoftheresearchprocess,thisexerciseprovidesyouwithanopportunitytoapplythatknowledgetoformulatearesearchproblemthatisofinteresttoyou.Asyouknow,selectingaresearchproblemisoneofthemostimportantaspectsof social research, so thisexercisewill, therefore,helpyou in formulatingyour researchproblembyraisingquestionsandissuesthatwillguideyoutoexaminecriticallyvariousfacetsandimplicationsofwhatyouareproposingtostudy.Theexerciseisdesignedtoprovideadirectionalframeworkthatguidesyouthroughtheproblemformulationpath.Keepinmindthatthequestionsandissuesraisedinthisexercisearenotprescriptivebutindicativeanddirectional,henceyouneedtobe critical and innovativewhileworking through them.Thinking through a researchproblemwithcarecanpreventatremendouswastageofhumanandfinancialresources.Aresearchproblemshouldbeclearlystatedandbespecificinnature.Thefeasibilityofthestudyintermsoftheavailabilityof
technicalexpertise, financesandtime,andin termsof itsrelevance,shouldbeconsideredthoroughlyat theproblem-formulationstage.Instudiesthatattempttoestablishacausalrelationshiporanassociation,theaccuracyofthemeasurementofindependent(cause) anddependent (effect) variables is of crucial importance and, hence, shouldbegiven serious consideration. If youhavealreadyselectedaproblem,youneednotgothroughthisprocess.
Startbyidentifyingabroadareayouareinterestedin.Forexample,ahealth,educationortreatmentprogramme;migration;patient
care;communityhealth;communityneeds;fostercare;or therelationshipbetweenunemploymentandstreetcrime.Chapter4 ofthisbookwillhelpyoutoworkthroughthisexercise.
StepI Selectabroadareaofstudythatinterestsyoufromwithinyouracademicdiscipline.
Havingselectedanarea,thenextstepisto‘dissect’itinordertoidentifyitsvariousaspectsandsubareas.Forexample,sayyourbroadareaof interest ismigration.Someaspectsorsubareasofmigrationare:
asocioeconomic–demographicprofileofimmigrants;reasonsforimmigration;problemsofimmigrants;servicesprovidedtoimmigrants;attitudesofimmigrantstowardsmigration;attitudesofhostcommunitiestowardsimmigrants;theextentofacculturationandassimilation;racialdiscriminationinthehostcountry.
Orperhapsyouareinterestedinstudyingapublichealthprogramme.Dissectitasfinelyaspossibleinordertoidentifytheaspectsthatcouldbestudied.Listthemastheycometoyou.Forexample:
asocioeconomic–demographicprofileofthetargetgroup;themorbidityandmortalitypatternsinacommunity;theextentandnatureofprogrammeutilisation;theeffectsofaprogrammeonacommunity;theeffectivenessofaparticularhealthpromotionstrategy.
Oryourinterestmaybeinstudyingdelinquents.Someaspectsofdelinquencyare:
delinquencyasrelatedtounemployment,brokenhomesorurbanisation;aprofileofdelinquents;reasonsfordelinquency;varioustherapeuticstrategies.
StepII ‘Dissect’thebroadareathatyouselectedinStepIintosubareasasdiscretelyandfinelyaspossible.Haveaone-person(withyourself)brainstormingsession.
1. ____________________2. ____________________3. ____________________4. ____________________
5. ____________________
Toinvestigateall thesesubareasisneitheradvisablenorfeasible.Selectonlythosesubareasthatwouldbepossible for you to studywithin the constraints of time, finance and expertise at yourdisposal.Onewaytoselectyoursubareaistostartwithaprocessofelimination:deletethoseareasyou are not very interested in. Towards the end it may become difficult but you need to keepeliminatinguntilyouhaveselectedasubarea(s)thatcanbemanagedwithinyourconstraints.Evenonesubareacanprovideyouwithavalidandexhaustivestudy.
StepIII From the above subareas, select a subarea or subareas in which you would like toconductyourstudy.
1. ____________________2. ____________________3. ____________________
StepIV Withineachchosensubarea,what researchquestionsdoyouhope toanswer? (Beasspecificaspossible.Youcanselectoneorasmanysubareasasyouwant.)
SubareaSpecificresearchquestionstobeanswered
1 (a)__________
(b)__________
(c)__________
(d)__________
(e)__________
SubareaSpecificresearchquestionstobeanswered
2 (a)__________
(b)__________
(c)__________
(d)__________
(e)__________
3 (a)__________
(b)__________
(c)__________
(d)__________
(e)__________
Theresearchquestionstobeansweredthroughthestudybecomethebasisofyourobjectives.Useaction-oriented words in the formulation of objectives. The main difference between researchquestionsandobjectives is thewaytheyarewritten.Questionsarewordedinquestionformandobjectivesarestatementsreferringtotheachievementofatask. Yourmainobjectiveshould indicate theoverall focusofyourstudyand thesubobjectives, itsspecificaspects.Subobjectives shouldbe listednumerically.Theyshouldbewordedclearlyandunambiguously.Makesureeachobjectivecontainsonlyoneaspectofthestudy.
StepV Onthebasisofyourresearchquestions,formulatethemainobjectiveandthesubobjectivesofyourstudy.
Mainobjective(themainfocusofyourstudy):
Subobjectives(specificaspectsofyourstudy):
1. ____________________2. ____________________3. ____________________4. ____________________5. ____________________
StepVI Carefullyconsiderthefollowingaspectsofyourstudy.
Nowyouhavedeveloped theobjectivesofyour study.Take some time to thinkabout them.Be
clearaboutwhattasksareinvolved,whattimeisrealisticallyrequiredandwhatskillsyouneedtodevelopinordertoconductyourstudy.Considertheseareascarefullyagain.
StepVII Double-check:
Ifyouranswertoanyofthesequestionsiseither‘no’or‘uncertain’,re-examinetheselectedaspectscarefullyandmaketheappropriatechangesinyourobjectives.What, in your opinion, is the relevanceof this study to theory andpractice?Howwill your study
contribute to the existing body of knowledge, help the practitioners in your profession and assist inprogrammedevelopmentandpolicyformulation?
Relevancetotheory:
Relevancetopractice:
Now thatyouhave formulatedyour researchproblem it is important toexamineyourobjective,researchquestionsandhypothesestoidentifyifyouhaveusedanyconceptsintheirformulation.Whenyouconvertconceptsintovariablesanunderstandingaboutvariablesplaysaveryimportantrole.Concepts arehighly subjective as theirunderstandingvaries fromperson topersonand, assuch,theymaynotbemeasurable.Anyconcept,perceptionorimaginationthatcanbemeasuredonanyoneofthefourmeasurementscales(nominal,ordinal,internalorratio)iscalledavariable.It is importantforconceptsusedinastudytobeoperationalisedinmeasurabletermssothat theextentofvariationinastudypopulation’sunderstandingofthemisreduced,ifnoteliminated.Atthisstage,whenyouhaveformulatedyourobjectives,itisimportantforyoutothinkhowyouwilloperationaliseanyconcepts
usedintheobjectives,researchquestionsorhypothesesformulated:whataretheirindicatorsandhowwilltheybemeasured?The following table suggests how you might operationalise the concept of ‘effectiveness’, in relation to a health education
programmeonAIDS.Itliststheindicatorsofeffectiveness(youcanhaveotherindicators)setsoutthevariablesthatmeasuretheindicatorsanddescribestheunitofmeasurementforthevariables.
This part of the exercise is designed to help you operationalise themajor concepts used in your study. Refer to Chapter 5 foradditionalinformationonvariables.
StepVIII Operationaliseyourconcepts.
Itisessentialtodevelopaworkingoroperationaldefinitionofyourstudypopulation.Forexample,who would you consider to be a patient, an immigrant, a youth, a psychologist, a teacher, adelinquentoraChristian?Workingdefinitionsplayacrucial role inavoidingambiguities in theselectionofasampleandhelpyoutonarrowyourstudypopulation.
StepIX Operationallydefineyourstudypopulation.
As discussed, some believe that one must have a hypothesis to undertake an investigation;however,intheauthor’sopinion,hypotheses,althoughtheybringclarity,specificityandfocustoaresearch problem, are not essential for a study. You can conduct a valid investigation withoutconstructingasingleformalhypothesis.Ontheotherhand,youcanconstructasmanyhypothesesasyouthinkappropriate.Inepidemiologicalstudies,tonarrowthefieldofinvestigation,onemustconstructahypothesisastotheprobablecauseoftheconditiontobeinvestigated.Ahypothesisisahunch,assumption,suspicion,assertionorideaaboutaphenomenon,relationshiporsituation,whichyouintend
toinvestigateinordertofindoutifyouareright.Ifitprovestoberight,yourassumptionwascorrect;hence,youprovethatyour
hypothesiswastrue.Otherwise,youconcludeyourhypothesistobefalse.Disprovingahypothesisisasimportantas,ormoreimportantthan,provingit.Asahypothesisisusuallyconstructedonthebasis
of what is commonly believed to be right, your disproving it might lead to something new that has been ignored by previousresearchers.Ahypothesisshouldbeconceptuallysimple,clearandspecific,andbecapableofverificationandbeingexpressedoperationally.Thereisaspecificwayofwritingahypothesis,withwhichyouneedtobefamiliar(refertoChapter6).
StepX Constructyourhypothesisorhypothesesforeachsubobjective/researchquestion.
Forqualitativestudies
Asmentionedearlier,thedifferenceinqualitativeandquantitativeresearchstudiesstartswiththeway you think about and formulate your research problem. In qualitative studies, the researchproblem is preferred to be broad, flexible and continuously formulated as the information iscollected.Intheprocessofdatacollection,ifyoufindsomethinginterestingrelatingtoyourbroadareaofstudy,youaddtheaspect(s)andchangethefocustoaccommodatethenewvision.Thisflexibilityisanimportantstrengthofqualitativeresearchbutitisalsoimportantthatyoudevelopaconceptualframeworkof
issue and questions for your study, as non-specificity about what you want to find out can often create problems for yourrespondents.Many do not feel comfortable or are not in a position to articulate themultiple aspects of an area without beingprompted.Forsituationslikethisitisimportantthatyouarefullypreparedwithaframeworkinmindforyourenquiry.Nodoubtyoucandevelopthisframeworkduringdatacollection,whiletalkingtoyourrespondents,butthismaycreateaproblemintermsofcompletenessandcomparabilitywiththeinformationobtainedduringtheearlyphaseofthestudy.Youcanminimisesomeoftheseproblemsbydevelopingaconceptualframeworkinadvance.Itisalsoimportantthatyoucommunicatewithrespondentsinspecifictermswithoutbiasorinfluencingtheirthinking.Remember, these are not the questions that youwill ask of your respondents. These are just reminders for raising issues or
questionsifnothingmuchisforthcomingfromarespondent.Inqualitativeresearchthefollowingwouldbeconsideredasbroadareasofinterest:
WhatdoesitmeantohaveachildwithADHDinthefamily?Howresilientisthiscommunity?Whatiscommunityresponsiveness?LivingwithHIV/AIDS.Howhasacommunitycopedafteramajorbushfireortsunami?
StepI Select a broad area of study that interests you or a question that you want to findanswerstothroughtheresearchstudy.
StepII
Havingselectedyourmainresearchquestionorbroadareaofstudy, listallquestionsthatyouwanttofindanswersto.Alsolistallissuesthatyouwanttodiscusswithyourrespondents. Your literature review, discussions with others and consultation withpotentialrespondentswillbeofimmensehelpatthisstage.
Questions:
Issues:
ExerciseII:Conceptualisingastudydesign
Quantitativestudies
ExerciseIhasbeendevelopedtohelpyoutodecidewhatyouwanttofindoutabout.Thenextstepistodecidehowtogoaboutit.Thisincludesdecidingonanoverallplanandselectingproceduresand methods that you propose to use during your research journey. The details of your plan,proceduresandmethodsbecomethecoreofyourstudydesign.A study design describes the design per se, that is the type of study design you propose to adopt; for example,whether the
proposedstudyiscross-sectional,correlationalorexperimental.Itshouldalsoprovidedetailsofthelogisticalproceduresrequiredforgatheringinformationfromthestudypopulation.Thisexercisehelpsyoutoputforwardyourargumentstojustifytheselectionofthedesignyouareproposingforyourstudy,criticallyexaminingitsstrengthsandweaknesses,andthusenablingyoutoselectthebestandworkablestudydesign.Theexercisealsochallengesyoutothinkthroughotherlogisticalproceduressuchasoutliningtheprocessof identifyingandcontactingyourstudypopulationandyourplantoobtaintherequiredinformationfromyourpotentialrespondents,thushelpingyoutodeveloptheroadmapforyourjourney.Forqualitativestudiestheprocessisthesamethoughitvariesincontent.Theissuesraisedinthisexercisewillhelpyoutoconceptualiseyourstudydesign.Chapter8detailsthevarioustypesofstudy
designinbothquantitativeandqualitativeresearchforyoutorefertowhileworkingthroughthisexercise.
A:Answerstothefollowingquestionswillhelpyoutodevelopyourstudydesign(StepII).
1. Isthedesignthatyouproposetoadopttoconductyourstudycross-sectional,longitudinal,experimentalorcomparativeinnature?Ifpossibledrawadiagramdepictingthedesign.
2. Whydidyouselectthisdesign?
3. What,inyouropinion,arethestrengthsofthisdesign?
4. Whataretheweaknessesandlimitationsofthisdesign?Weaknesses:
Limitations:
5. Whoconstitutesyourstudypopulation?
6. Willyoubeabletoidentifyeachrespondentinyourstudypopulation?Yes No
6(a)Ifyes,howwilltheybeidentified?
6(b)Ifno,howdoyouplantogetintouchwiththem?
7. Doyouplantoselectasample?
Yes No
7(a)Ineithercase,explainthereasonsforyourdecision.
8. Howwillyoucollectdatafromyourrespondents(e.g.interview,questionnaire)?
8(a)Whydidyouselectthismethodofdatacollection?
8(b)What,inyouropinion,areitsstrengthsandweaknesses?Strengths:
Weaknesses:
8(c)Ifyouareinterviewing,wherewilltheinterviewsbeheld?
8(d)Ifyouareusingmailedquestionnaires:
(i)Fromwherewillyouobtaintheaddressesofpotentialrespondents?
(ii)Areyouplanningtoencloseaself-addressedstampedenvelopewiththequestionnaires?
Yes No
(iii)Inthecaseofalowresponserate,willyousendareminder?
Yes No
(iv)Iftherearequeries,howshouldrespondentsgetintouchwithyou?
B: On the basis of the above information, describe your study design. (For further guidance, consultChapter8.)
Forqualitativestudies
A:Answerstothefollowingquestionswillhelpyouindevelopingaroadmapforyourresearchjourney.
1. Inwhichgeographicalarea,community,grouporpopulationgroupwouldyouliketoundertakeyourstudy?
2. Howdoyouplantogetentryintothearea,communityorgroup?Whichnetwork,ifany,areyouplanningtouse?
3. Whydidyouselectthisgroup?
4. Fromwhomwillyougathertherequiredinformation?(Whowillbeyourrespondents?)
5. Ifyouaregatheringinformationfromsecondarysources,haveyoucheckedtheiravailability?
Yes No Notapplicable
6. Haveyoucheckedtheavailabilityoftherequiredinformationinthem?
Yes No Notapplicable
7. Ifyouaregatheringinformationfromindividuals,howmanywillyoucontact?
8. Whatwillbethebasisofselectionoftheseindividuals?
9. Howwillyoucollecttherequiredinformation?Listallmethodsthatyouplantouse.
ExerciseIII:Developingaresearchinstrument
Theconstructionofaresearchinstrumentisthefirstpracticalstepinoperationalisingyourstudy.Itisanimportantaspectofyourresearchas itconstitutes theinput; thequalityofyouroutput(thefindingsandconclusions)isentirelydependentuponthequalityandappropriatenessofyourinput– the research instrument. Items in a research instrument are questions asked of respondents.Responses to these questions become the raw data that is processed to find answers to yourresearch questions.The famous saying about computers, ‘garbage in, garbage out’, also equallyappliestotheresearchinstrument.Toalargeextent,thevalidityofthefindingsdependsuponthequalityoftherawdatawhich,inturn,dependsupontheresearchinstrumentyouhaveused.Ifthelatterisvalidandreliable,thefindingsshouldalsobevalidandreliable.Thequalityofaresearchinstrumentlargelydependsuponyourexperienceinresearch.Itisimportantforabeginnertofollowthe
suggestedstepsoutlinedinChapter9.
Forquantitativestudies
Quantitativeresearchisstructuredandpredeterminedintermsofwhatyouwanttofindoutaboutand how. As a part of this operational step, you need to decide what questions to ask of yourrespondents,thewordingyouaregoingtouseandtheorderinwhichthequestionswillbeasked.Thisexerciseisdesignedtohelptodevelopskillsinconstructinganinstrument.One of the ways to formulate the questions that are going to constitute your research instrument is by examining each
subobjective/research question/hypothesis you have developed for your study, specifying for each the information you require,identifyingthevariablesthatareneeded,andthenbyformulatingquestionstobeaskedofyourrespondents.Thewordingofyourquestionsshouldbesimpleandwithoutambiguities.Donotaskleadingquestionsorquestionsbasedupon
presumptions.Double-barrelledquestionsshouldbeavoided.Thepre-testofaresearchinstrumentisanintegralpartofinstrumentconstruction.Asarule,thepre-testshouldnotbecarriedout
onyoursamplebutonasimilarpopulation.
StepIOn a separate piece of paper, draw a table as shown below, then list all your sub-objectives/researchquestions/hypothesesinthefirstcolumnandworkthroughtheothercolumnslistingtherequiredinformation.
StepII
Formulate the questions,* preferably on a separate piece of paper, giving particularattentiontotheirwordingandorder.Inyourownmindyoumustexaminetherelevanceand justification of each question in relation to the objectives of your study. If youcannot relate the relevance and justification of a question to the objectives of yourstudy,itshouldbediscarded.
StepIII Ifyouaredevelopingaquestionnaire,incorporateinteractivestatementsatappropriateplaces.
StepIVAfterdevelopingthefirstdraftofyourresearchinstrument,answerthequestionsyourself;thatis,interviewyourselforcompletethequestionnaire.Youneedtoimaginethatyouareamemberofthestudypopulationwhowillbeaskedthesequestionsorrequestedtocompletethequestionnaire.Ifyoufinditdifficulttoansweraquestion,re-examineit.
StepV
Onceyouaresatisfiedwiththeresearchinstrument,pre-testitwithafewrespondentsfromapopulationsimilartothe one you are going to study.The purpose of the pre-test/field test is not to obtain information but to uncoverproblemswiththeinstrument.Iftheinstrumentisaninterviewschedule,interviewthepre-testrespondentstofindoutiftheyunderstandthequestions.Ifaquestionisnotunderstood,findoutwhattherespondentdidnotunderstand.If the same problem is identified by more than one respondent, change the wording. If your instrument is aquestionnaire,askthepre-testrespondentstogothroughthequestionswiththeaimofidentifyinganyquestionsthataredifficulttounderstand.Discusstheproblemsthattheyhadinunderstandingorinterpretingaquestion.Inlightofthesediscussions,ifnecessary,changethewordingofquestionswithwhichpre-testrespondentshavedifficulties.
StepVI
Having pre-tested and, if necessary, amended the instrument, take a piece of paper and draw a table with twocolumns.Inthefirstcolumnwriteeachsubobjective,researchquestionandhypothesisseparately,andintheother,writethequestionnumber(s)thatprovideinformationfortheseobjectives,researchquestionsorhypotheses.Inotherwords,makeeachquestionmatchtheobjectiveforwhichitprovidesinformation.Ifaquestioncannotbelinkedtoaspecificobjective,researchquestionorhypothesis,examinewhyitwasincluded.
StepVII Preparethefinaldraftofyourresearchinstrument.Ifyouplantouseacomputerfordataanalysis,youmayprovidespaceontheresearchinstrumentforcodingthedata.
Forqualitativestudies
Ifyouaredoingaqualitativestudy,youdonotneedtodevelopalistofquestions.However,itisimportantthatyouconstructaframeworkoftheissuesthatyouthinkyoushouldcovertoachievethe objectives of your study. This interview guide or conceptual framework will help you tocontinuewithyourinterviewsifnothingmuchisforthcomingfromyourrespondents.Youraimistoletarespondentbringouttheissues,butthisframeworkisreadyincasethatdoesnothappen.ConsultChapter9fordevelopingaconceptualframework.
Write,inapointform,theissuesthatyouthinkyouwanttodiscusswithyourrespondents.MostofityouhavealreadydoneasapartofExerciseI.
ExerciseIV:Selectingasample
Theaccuracyofwhatyoufindthroughyourresearchendeavour,inadditiontomanyotherthings,dependsuponthewayyouselectyoursample.Theunderlyingpremiseinsamplingisthatasmallnumberofunits,ifselectedcorrectly,canprovide,toasufficientlyhighdegree
ofprobability,reasonablyaccurateinsightintowhatishappeninginthestudypopulation.Fordetailsonsamplingdesigns,refertoChapter12.
Forquantitativestudies
The basic objective of a sampling design in quantitative research is to minimise, within thelimitationofcost,anydifferencebetweenthevaluesandestimatesobtainedfromyoursampleandthoseprevalentinthestudypopulation.Samplingtheory,inquantitativeresearch,isthusguidedbytwoprinciples:
1. theavoidanceofbiasintheselectionofasample;2. theattainmentofmaximumprecisionforagivenoutlayofresources.
In quantitative research you can select any of the probability or a non-probability sample design. Both have advantages anddisadvantagesandbothareappropriateforcertainsituations.Butwhateversamplingdesignyouchoose,makesureyoutakestepstoavoidintroducingyourbias.Whenselectingasampleinquantitativestudiesyouneedtodecideontwothings:thesamplesizeyouplantoselect;andhowtoselecttherequiredsamplingunits.Youalsoneedtothinkaboutyourreasonsfordecidingthesizeandchoosingthesamplingstrategy.
Thisexerciseisdesignedforyoutothinkthroughtheissueswhichareimportantinhelpingyoutodevelopyoursamplingstrategy.
StepI Answerthefollowingaboutyoursamplingdesign.
1. Whatisthetotalsizeofyourstudypopulation?__________Unknown
2. Doyouwanttoselectasample?Yes No2(a)Ifyes,whatwillyoursamplesizebe?2(b)Whatareyourreasonsforchoosingthissamplesize?__________
3. Howwillyouselectyoursample?(Whatsamplingdesignareyouproposing?)4. Whydidyouselectthissamplingdesign?(Whatareitsstrengths?)5. Whatarethelimitationsofthisdesign?
StepII Onthebasisoftheanswerstotheabovequestions,writeaboutyoursamplingdesign,detailingtheprocessandyourjustificationforusingit.
Forqualitativestudies
Inqualitativeresearchyouraimisnot toselectarandomorunbiasedsamplebutonewhichcanprovideyou,asfaraspossible,withthedetailed,accurateandcompleteinformationthatyouarelooking for. Hence, you are dominantly guided by your judgement in the selection of yourrespondents.In qualitative research you can only use non-probability designs but you are not guided by the sample size. The numbers of
peopleyouaregoingtocontactdependupontheattainmentofthedatasaturationpointduringthedatacollectionprocess.Youalsoneedtodecidewhoaregoingtobeyourrespondentsandhowtheyaregoingtobeidentified.Youneedtothinkabout
thedeterminantsonwhichyouaregoingtobaseyourjudgementastothesuitabilityofyourrespondents.Answerstothefollowingquestionswillhelpyoutothinkthroughtheissuesyouarelikelytofacewhiledevelopingasampling
strategyforyourstudy.
A:Whatfactorswouldyoukeepinmindwhenselectingarespondent?
B:Howwouldyouidentifyyourpotentialrespondents?
ExerciseV:Developingaframeofanalysis
Forbothquantitativeandqualitativestudiesingeneraltermsdescribethestrategyyouplantousefordataanalysis.Decidewhetherthedatawillbeanalysedmanuallyorbycomputer.Forcomputeranalysisyouneedtoidentifytheprogramyouplantouse.RefertoChapter15fordetails.
Forquantitativestudies
Youshouldalsospecifythetypeofanalysisyouplantocarryout–thatis,frequencydistributions,cross-tabulations, regression analysis or analysis of variance. It is also important to planwhichvariableswillbesubjectedtowhichtypeofstatisticalprocedure.Ifyouhaveusedcertainconceptsinyourstudy,howwilltheseconceptsbeoperation-alised?Forexample,ifyouweremeasuring
theeffectivenessofahealthprogramme,howwouldtheresponsestothevariousquestions,designedtofindouttheeffectiveness,becombinedtoascertaineffectiveness?Keepinmindthatwhenyouactuallycarryoutdataanalysis,itisonlynaturalthatyouwilldevelopnew ideasonhow to improve the analysis of data.You should feel free to change the frameof analysiswhen actuallyanalysingdataifyousodesire.Thisexerciseisaroughguideforyoutostartthinkingaboutanalysingyourrawdata.
Thinkthroughthefollowingissues:
1. Ifyouareplanningtouseacomputerfordataanalysis,whatsoftwarewillyouuse?
2. Whichvariableswillyousubjecttofrequencydistributionanalysis?
3. Whichvariableswillbecross-tabulated?
4. Whatvariableswillbesubjectedtowhichstatisticalprocedures(e.g.regressionanalysis,ANOVA,factoranalysis)?
5. Howdoyouplantooperationaliseorconstructthemainconceptsthroughcombiningresponsestodifferentquestions(e.g.satisfactionindex,effectiveness)?
Forqualitativestudies
Forqualitativestudiesitisalsoimportantforyoutospecifythetypeofanalysisyouaregoingtohave. If it is a description or narration of an event, episode, situation or instance, you shouldoutlinehowitisgoingtobestructured.Ifyouaregoingtoidentifythemainthemes,youshouldspecifyhowyouareplanningtoanalysethecontentstoidentifythem.Keepinmindthatasyougothroughtheanalysisyouwillgetmanynewideaswhichyouwillneedtoincorporate.
1. Ifyouareplanningtouseacomputerfordataanalysis,whatsoftwarewillyouuse?
2. Howareyougoingtoidentifythemainthemesthatemergedfromyourfieldnotes,in-depthinterviewsoranyothersourcethatyouused?
3. Areyougoingtoquantifythesethemes?Ifyes,how?
ExerciseVI:Developinganoutlineofthechapters
Althougheachoperationalstep is important, inawaywriting thereport is themostcrucialas ittellsothersabouttheoutcomeofyourstudy:itistheoutcomeofthehardworkyouhaveputintoyourstudyandistheonlythingvisibletoreaders.Hence,eventhemostvaluableworkcouldbelostifyourreportisnotwellwritten.The quality of your report depends upon many things: your writing skills; the clarity of your thoughts and their logical
expression;yourknowledgeofthesubject;andyourexperienceinresearchwriting.
Developinganoutline for the structureof the report is extremelyuseful.Asabeginner it is important thatyou think throughcarefullythecontentsofyourreport,organisethemaroundthemainthemesofyourstudy,andensurethatthevariousaspectsofathemearewellintegratedandfollowalogicalprogression.Thisexerciseisdesignedtohelpyoutoorganiseyourthoughtswithrespecttowritingyourresearchreportwhetheryourstudyis
quantitativeorqualitative.Youshould,asfaraspossible,attempttoplacethevariousaspectsofyourreportunderchapterheadings,eveniftheseareverytentative.Forthisexercisedevelopheadingsforyourchaptersandthenlisttheirtentativecontents.Keepinmind that these are likely to change as you startwriting the actual report. Though they can sometimes completely change, thisexercisewillstillbeveryrewardingandmayprovideyouwithvaluableguidanceinorganisingyourthoughtsandwriting.ConsultChapter17formoredetails.
1. Whatarethemainthemesofyourstudy?
2. Developchapterheadingsunderwhichtheabovethemeswillbeorganisedinwritingyourreport.
3. Developanoutlineofeachchapterwhichbrieflydescribeswhatyouaregoingtowritewithineachchapter.
Glossary
100percentbarchart:The100percentbarchartisverysimilartothestackedbarchart.Theonlydifferenceisthatintheformerthesubcategoriesofavariableforaparticularbartotal100percentandeachbarisslicedintoportionsinrelationtotheirproportionoutof100.
Accidentalsampling,asquotasampling,isbaseduponyourconvenienceinaccessingthesamplingpopulation.Whereasquotasamplingattemptstoincludepeoplepossessinganobvious/visiblecharacteristic,accidentalsamplingmakesnosuchattempt.Anypersonthatyoucomeacrosscanbecontactedforparticipationinyourstudy.Youstopcollectingdatawhenyoureachtherequirednumberofrespondentsyoudecidedtohaveinyoursample.
Actionresearch,incommonwithparticipatoryresearchandcollaborativeenquiry,isbaseduponaphilosophyofcommunitydevelopmentthatseekstheinvolvementofcommunitymembersinplanning,undertaking,developingandimplementingresearchandprogrammeagendas.Researchisameanstoactiontodealwithaproblemoranissueconfrontingagrouporcommunity.Itfollowsacyclicalprocessthatisusedtoidentifytheissues,developstrategiesandimplementtheprogrammestodealwiththemandthenagainassessingstrategiesinlightoftheissues.
Activevariable:Instudiesthatseektoestablishcausalityorassociationtherearevariablesthatcanbechanged,controlledandmanipulatedeitherbyaresearcherorbysomeoneelse.Suchvariablesarecalledactivevariables.
After-onlydesign:Inanafter-onlydesigntheresearcherknowsthatapopulationisbeing,orhasbeen,exposedtoaninterventionandwishestostudyitsimpactonthepopulation.Inthisdesign,baselineinformation(pre-testorbeforeobservation)isusually‘constructed’eitheronthebasisofrespondents’recallofthesituationbeforetheintervention,orfrominformationavailableinexistingrecords,i.e.secondarysources.
Alternatehypothesis:Theformulationofanalternatehypothesisisaconventioninscientificcircles.Itsmainfunctionistospecifyexplicitlytherelationshipthatwillbeconsideredastrueincasetheresearchhypothesisprovestobewrong.Inaway,analternatehypothesisistheoppositeoftheresearchhypothesis.
Ambiguousquestion:Anambiguousquestionisonethatcontainsmorethanonemeaningandthatcanbeinterpreteddifferentlybydifferentrespondents.
Appliedresearch:Mostresearchinthesocialsciencesisappliedinnature.Appliedresearchisonewhereresearchtechniques,proceduresandmethodsthatformthebodyofresearchmethodologyareappliedtocollectinformationaboutvariousaspectsofasituation,issue,problemorphenomenonsothattheinformationgatheredcanbeutilisedforotherpurposessuchaspolicyformulation,programmedevelopment,programmemodificationandevaluation,enhancementoftheunderstandingaboutaphenomenon,establishingcausalityandoutcomes,identifyingneedsanddevelopingstrategies.
Areachart:Forvariablesmeasuredonanintervaloraratioscale,informationaboutthesub-categoriesofavariablecanalsobepresentedintheformofanareachart.Itisplottedinthesamewayasalinediagramwiththeareaundereachlineshadedtohighlightthemagnitudeofthesubcategoryinrelationtoothersubcategories.Thusanareachartdisplaystheareaunderthecurveinrelationtothesubcategoriesofavariable.
Attitudinalscales:Thosescalesthataredesignedtomeasureattitudestowardsanissuearecalledattitudinalscales.Inthesocialsciencestherearethreetypesofscale:thesummatedratingscale(Likertscale),theequal-appearingintervalscale(Thurstonescale)andthecumulativescale(Guttmanscale).
Attitudinalscore:Anumberthatyoucalculatehavingassignedanumericalvaluetotheresponsegivenbyarespondenttoanattitudinalstatementorquestion.Differentattitudescaleshavedifferentwaysofcalculatingtheattitudinalscore.
Attitudinalvalue:Anattitudinalscalecomprisesmanystatementsreflectingattitudestowardsanissue.Theextenttowhicheachstatementreflectsthisattitudevariesfromstatementtostatement.Somestatementsaremoreimportantindeterminingtheattitudethanothers.Theattitudinalvalueofastatementreferstotheweightcalculatedorgiventoastatementtoreflectitssignificanceinreflectingtheattitude:thegreaterthesignificanceorextent,thegreatertheattitudinalvalueorweight.
Attributevariables:Thosevariablesthatcannotbemanipulated,changedorcontrolled,andthatreflectthecharacteristicsofthestudypopulation.Forexample,age,gender,educationandincome.
Barchart:Thebarchartordiagramisoneofthewaysofgraphicallydisplayingcategoricaldata.Abarchartisidenticaltoahistogram,exceptthatinabarcharttherectanglesrepresentingthevariousfrequenciesarespaced,thusindicatingthatthedataiscategorical.Thebardiagramisusedforvariablesmeasuredonnominalorordinalscales.
Before-and-afterstudies:Abefore-and-afterdesigncanbedescribedastwosetsofcross-sectionaldatacollectionpointsonthesamepopulationtofindoutthechangeinaphenomenonorvariable(s)betweentwopointsintime.Thechangeismeasuredbycomparingthedifferenceinthephenomenonorvariable(s)betweenbeforeandafterobservations.
Biasisadeliberateattempteithertoconcealorhighlightsomethingthatyoufoundinyourresearchortousedeliberatelyaprocedureormethodthatyouknowisnotappropriatebutwillprovideinformationthatyouarelookingforbecauseyouhaveavestedinterestinit.
Blindstudies:Inablindstudy,thestudypopulationdoesnotknowwhetheritisgettingrealorfaketreatmentorwhichtreatmentmodalityinthecaseofcomparativestudies.Themainobjectiveofdesigningablindstudyistoisolatetheplaceboeffect.
Casestudy:Thecasestudydesignisbasedupontheassumptionthatthecasebeingstudiedisatypicalofcasesofacertaintypeandthereforeasinglecasecanprovideinsightintotheeventsandsituationsprevalentinagroupfromwherethecasehasbeendrawn.Inacasestudydesignthe‘case’youselectbecomesthebasisofathorough,holisticandin-depthexplorationoftheaspect(s)thatyouwanttofindoutabout.Itisanapproachinwhichaparticularinstanceorafewcarefullyselectedcasesarestudiedintensively.Tobecalledacasestudyitisimportanttotreatthetotalstudypopulationasoneentity.Itisoneoftheimportantstudydesignsinqualitativeresearch.
Categoricalvariablesarethosewheretheunitofmeasurementisintheformofcategories.Onthebasisofpresenceorabsenceofacharacteristic,avariableisplacedinacategory.Thereisnomeasurementofthecharacteristicsassuch.Intermsofmeasurementscalessuchvariablesaremeasuredonnominalorordinalscales.Rich/poor,high/low,hot/coldareexamplesofcategoricalvariables.
Chancevariable:Instudyingcausalityorassociationtherearetimeswhenthemoodofarespondentorthewordingofaquestioncanaffectthereplygivenbytherespondentwhenaskedagaininthepost-test.Thereisnosystematicpatternintermsofthischange.Suchvariablesarecalledchanceorrandomvariables.
Closedquestion:Inaclosedquestionthepossibleanswersaresetoutinthequestionnaireorinterviewscheduleandtherespondentortheinvestigatorticksthecategorythatbestdescribearespondent’sanswer.
Clustersampling:Clustersamplingisbasedontheabilityoftheresearchertodivideasamplingpopulationintogroups(baseduponavisibleoreasilyidentifiablecharacteristics),calledclusters,andthenselectelementsfromeachclusterusingtheSRStechnique.Clusterscanbeformedonthebasisofgeographicalproximityoracommoncharacteristicthathasacorrelationwiththemainvariableofthestudy(asinstratifiedsampling).Dependingonthelevelofclustering,sometimessamplingmaybedoneatdifferentlevels.Theselevelsconstitutethedifferentstages(single,doubleormultiple)ofclustering.
Code:Thenumericalvaluethatisassignedtoaresponseatthetimeofanalysingthedata.
Codebook:Alistingofasetofnumericalvalues(setofrules)thatyoudecidedtoassigntoanswersobtainedfromrespondentsinresponsetoeachquestioniscalledacodebook.
Coding:Theprocessofassigningnumericalvaluestodifferentcategoriesofresponsestoaquestionforthepurposeofanalysingthemiscalledcoding.
Cohortstudiesarebasedupontheexistenceofacommoncharacteristicsuchasyearofbirth,graduationormarriage,withinasubgroupofapopulationthatyouwanttostudy.Peoplewiththecommoncharacteristicsarestudiedoveraperiodoftimetocollecttheinformationofinteresttoyou.Studiescouldcoverfertilitybehaviourofwomenbornin1986orcareerpathsof1990graduatesfromamedicalschool,forinstance.Cohortstudieslookatthetrendsoveralongperiodoftimeandcollectdatafromthesamegroupofpeople.
Collaborativeenquiryisanothernameforparticipatoryresearchthatadvocatesaclosecollaborationbetweentheresearcherandtheresearchparticipants.
Columnpercentagesarecalculatedfromthetotalofallthesubcategoriesofonevariablethataredisplayedalongacolumnindifferentrows.
Communitydiscussionforum:Acommunitydiscussionforumisaqualitativestrategydesignedtofindopinions,attitudes,ideasofacommunitywithregardtocommunityissuesandproblems.Itisoneoftheverycommonwaysofseekingacommunity’sparticipationindecidingaboutissuesofconcerntoit.
Comparativestudydesign:Sometimesyouseektocomparetheeffectivenessofdifferenttreatment
modalities.Insuchsituationsacomparativedesignisused.Withacomparativedesign,aswithmostotherdesigns,astudycanbecarriedouteitherasanexperimentornon-experiment.Inthecomparativeexperimentaldesign,thestudypopulationisdividedintothesamenumberofgroupsasthenumberoftreatmentstobetested.Foreachgroupthebaselinewithrespecttothedependentvariableisestablished.Thedifferenttreatmentmodalitiesarethenintroducedtothedifferentgroups.Afteracertainperiod,whenitisassumedthatthetreatmentmodelshavehadtheireffect,the‘after’observationiscarriedouttoascertainchangesinthedependentvariable.
Concept:Indefiningaresearchproblemorthestudypopulationyoumayusecertainwordsthatassucharedifficulttomeasureand/ortheunderstandingofwhichmayvaryfrompersontoperson.Thesewordsarecalledconcepts.Inordertomeasurethemtheyneedtobeconvertedintoindicators(notalways)andthenvariables.Wordslikesatisfaction,impact,young,old,happyareconceptsastheirunderstandingwouldvaryfrompersontoperson.
Conceptualframework:Aconceptualframeworkstemsfromthetheoreticalframeworkandconcentrates,usually,ononesectionofthattheoreticalframeworkwhichbecomesthebasisofyourstudy.Thelatterconsistsofthetheoriesorissuesinwhichyourstudyisembedded,whereastheformerdescribestheaspectsyouselectedfromthetheoreticalframeworktobecomethebasisofyourresearchenquiry.Theconceptualframeworkisthebasisofyourresearchproblem.
Concurrentvalidity:Whenyouinvestigatehowgoodaresearchinstrumentisbycomparingitwithsomeobservablecriterionorcrediblefindings,thisiscalledconcurrentvalidity.Itiscomparingthefindingsofyourinstrumentwiththosefoundbyanotherwhichiswellaccepted.Concurrentvalidityisjudgedbyhowwellaninstrumentcompareswithasecondassessmentdoneconcurrently.
Conditioningeffect:Thisdescribesasituationwhere,ifthesamerespondentsarecontactedfrequently,theybegintoknowwhatisexpectedofthemandmayrespondtoquestionswithoutthought,ortheymayloseinterestintheenquiry,withthesameresult.Thissituation’seffectonthequalityoftheanswersisknownastheconditioningeffect.
Confirmabilityreferstothedegreetowhichtheresultsobtainedthroughqualitativeresearchcouldbeconfirmedorcorroboratedbyothers.Confirmabilityinqualitativeresearchissimilartoreliabilityinquantitativeresearch.
Constantvariable:Whenavariablecanhaveonlyonecategoryorvalue,forexampletaxi,treeandwater,itisknownasaconstantvariable.
Constructvalidityisamoresophisticatedtechniqueforestablishingthevalidityofaninstrument.Constructvalidityisbaseduponstatisticalprocedures.Itisdeterminedbyascertainingthecontributionofeachconstructtothetotalvarianceobservedinaphenomenon.
Consumer-orientedevaluation:Thecorephilosophyofthisevaluationrestsontheassumptionthatassessmentofthevalueormeritofanintervention–includingitseffectiveness,outcomes,impactandrelevance–shouldbejudgedfromtheperspectiveoftheconsumer.Consumers,accordingtothisphilosophy,arethebestpeopletomakeajudgementontheseaspects.Anevaluationdonewithintheframeworkofthisphilosophyisknownasconsumer-orientedevaluationorclient-centredevaluation.
Contentanalysisisoneofthemainmethodsofanalysingqualitativedata.Itistheprocessofanalysing
thecontentsofinterviewsorobservationalfieldnotesinordertoidentifythemainthemesthatemergefromtheresponsesgivenbyyourrespondentsortheobservationnotesmadebyyouasaresearcher.
Contentvalidity:Inadditiontolinkingeachquestionwiththeobjectivesofastudyasapartofestablishingthefacevalidity,itisalsoimportanttoexaminewhetherthequestionsoritemshavecoveredalltheareasyouwantedtocoverinthestudy.Examiningquestionsofaresearchinstrumenttoestablishtheextentofcoverageofareasunderstudyiscalledcontentvalidityoftheinstrument.
Continuousvariableshavecontinuityintheirunitofmeasurement;forexampleage,incomeandattitudescore.Theycantakeonanyvalueofthescaleonwhichtheyaremeasured.Agecanbemeasuredinyears,monthsanddays.Similarly,incomecanbemeasuredindollarsandcents.
Controldesign:Inexperimentalstudiesthataimtomeasuretheimpactofanintervention,itisimportanttomeasurethechangeinthedependentvariablethatisattributedtotheextraneousandchancevariables.Toquantifytheimpactofthesesetsofvariablesanothercomparablegroupisselectedthatisnotsubjectedtotheintervention.Studydesignswhereyouhaveacontrolgrouptoisolatetheimpactofextraneousandchangevariablesarecalledcontroldesignstudies.
Controlgroup:Thegroupinanexperimentalstudywhichisnotexposedtotheexperimentalinterventioniscalledacontrolgroup.Thesolepurposeofthecontrolgroupistomeasuretheimpactofextraneousandchancevariablesonthedependentvariable.
Correlationalstudies:Studieswhichareprimarilydesignedtoinvestigatewhetherornotthereisarelationshipbetweentwoormorevariablesarecalledcorrelationalstudies.
Cost–benefitevaluation:Thecentralaimofacost–benefitevaluationistoputapricetagonaninterventioninrelationtoitsbenefits.
Cost-effectivenessevaluation:Thecentralaimofacost-effectivenessevaluationistoputapricetagonaninterventioninrelationtoitseffectiveness.
Credibilityinqualitativeresearchisparalleltointernalvalidityinquantitativeresearchandreferstoasituationwheretheresultsobtainedthroughqualitativeresearchareagreeabletotheparticipantsoftheresearch.Itisjudgedbytheextentofrespondentconcordancewherebyyoutakeyourfindingstothosewhoparticipatedinyourresearchforconfirmation,congruence,validationandapproval:thehighertheoutcomeofthese,thehigherthecredibility(validity)ofthestudy.
Cross-overcomparativeexperimentaldesign:Inthecross-overdesign,alsocalledtheABABdesign,twogroupsareformed,theinterventionisintroducedtooneofthemand,afteracertainperiod,theimpactofthisinterventionismeasured.Thentheinterventionsare‘crossedover’;thatis,theexperimentalgroupbecomesthecontrolandviceversa.
Cross-sectionalstudies,alsoknownasone-shotorstatusstudies,arethemostcommonlyuseddesigninthesocialsciences.Thisdesignisbestsuitedtostudiesaimedatfindingouttheprevalenceofaphenomenon,situation,problem,attitudeorissue,bytakingacross-sectionofthepopulation.Theyareusefulinobtaininganoverall‘picture’asitstandsatthetimeofthestudy.
Cross-tabulationisastatisticalprocedurethatanalysestwovariables,usuallyindependentand
dependentorattributeanddependent,todetermineifthereisarelationshipbetweenthem.Thesubcategoriesofboththevariablesarecross-tabulatedtoascertainifarelationshipexistsbetweenthem.
Cumulativefrequencypolygon:Thecumulativefrequencypolygonorcumulativefrequencycurveisdrawnonthebasisofcumulativefrequencies.Themaindifferencebetweenafrequencypolygonandacumulativefrequencypolygonisthattheformerisdrawnbyjoiningthemidpointsoftheintervals,whereasthelatterisdrawnbyjoiningtheendpointsoftheintervalsbecausecumulativefrequenciesinterpretdatainrelationtotheupperlimitofaninterval.
Dependabilityinqualitativeresearchisverysimilartotheconceptofreliabilityinquantitativeresearch.Itisconcernedwithwhetherwewouldobtainthesameresultsifwecouldobservethesamethingtwice:thegreaterthesimilarityintworesults,thegreaterthedependability.
Dependentvariable:Whenestablishingcausalitythroughastudy,thevariableassumedtobethecauseiscalledanindependentvariableandthevariablesinwhichitproduceschangesarecalledthedependentvariables.Adependentvariableisdependentupontheindependentvariableanditisassumedtobebecauseofthechanges.
Descriptivestudies:Astudyinwhichthemainfocusisondescription,ratherthanexaminingrelationshipsorassociations,isclassifiedasadescriptivestudy.Adescriptivestudyattemptssystematicallytodescribeasituation,problem,phenomenon,serviceorprogramme,orprovidesinformationabout,say,thelivingconditionsofacommunity,ordescribesattitudestowardsanissue.
Dichotomousvariable:Whenavariablecanhaveonlytwocategoriesasinmale/female,yes/no,good/bad,head/tail,up/downandrich/poor,itisknownasadichotomousvariable.
Disproportionatestratifiedsampling:Whenselectingastratifiedsampleifyouselectanequalnumberofelementsfromeachstratumwithoutgivinganyconsiderationtoitssizeinthestudypopulation,theprocessiscalleddisproportionatestratifiedsampling.
Double-barrelledquestion:Adouble-barrelledquestionisaquestionwithinaquestion.
Double-blindstudies:Theconceptofadouble-blindstudyisverysimilartothatofablindstudyexceptthatitalsotriestoeliminateresearcherbiasbynotdisclosingtotheresearchertheidentitiesofexperimental,comparativeandplacebogroups.Inadouble-blindstudyneithertheresearchernorthestudyparticipantsknowwhichstudyparticipantsarereceivingreal,placeboorotherformsofinterventions.Thispreventsthepossibilityofintroducingbiasbytheresearcher.
Double-controlstudies:Althoughthecontrolgroupdesignhelpsyoutoquantifytheimpactthatcanbeattributedtoextraneousvariables,itdoesnotseparateoutothereffectsthatmaybeduetotheresearchinstrument(suchasthereactiveeffect)orrespondents(suchasthematurationorregressioneffects,orplaceboeffect).Whenyouneedtoidentifyandseparateouttheseeffects,adouble-controldesignisrequired.Inadouble-controlstudy,youhavetwocontrolgroupsinsteadofone.Toquantify,say,thereactiveeffectofaninstrument,youexcludeoneofthecontrolgroupsfromthe‘before’observation.
Editingconsistsofscrutinisingthecompletedresearchinstrumentstoidentifyandminimise,asfaraspossible,errors,incompleteness,misclassificationandgapsintheinformationobtainedfromrespondents.
Elevationeffect:Someobserverswhenusingascaletorecordanobservationmayprefertousecertainsection(s)ofthescaleinthesamewaythatsometeachersarestrictmarkersandothersarenot.Whenobservershaveatendencytouseaparticularpart(s)ofascaleinrecordinganinteraction,thisphenomenonisknownastheelevationeffect.
Errorofcentraltendency:Whenusingscalesinassessmentsorobservations,unlessanobserverisextremelyconfidentofhis/herabilitytoassessaninteraction,s/hemaytendtoavoidtheextremepositionsonthescale,usingmostlythecentralpart.Theerrorthistendencycreatesiscalledtheerrorofcentraltendency.
Ethicalpractice:Professionalpracticeundertakeninaccordancewiththeprinciplesofacceptedcodesofconductforagivenprofessionorgroup.
Evaluationisaprocessthatisguidedbyresearchprinciplesforreviewinganinterventionorprogrammeinordertomakeinformeddecisionsaboutitsdesirabilityand/oridentifyingchangestoenhanceitsefficiencyandeffectiveness.
Evaluationforplanningaddressestheissueofestablishingtheneedforaprogrammeorintervention.
Evidence-basedpractice:Aservicedeliverysystemthatisbaseduponresearchevidenceastoitseffectiveness;aserviceprovider’sclinicaljudgementastoitssuitabilityandappropriatenessforaclient;andaclient’spreferenceastoitsacceptance.
Experimentalgroup:Anexperimentalgroupisonethatisexposedtotheinterventionbeingtestedtostudyitseffects.
Experimentalstudies:Instudyingcausality,whenaresearcherorsomeoneelseintroducestheinterventionthatisassumedtobethe‘cause’ofchangeandwaitsuntilithasproduced–orhasbeengivensufficienttimetoproduce–thechange,theninstudieslikethisaresearcherstartswiththecauseandwaitstoobserveitseffects.Suchtypesofstudiesarecalledexperimentalstudies.
Expertsamplingistheselectionofpeoplewithdemonstratedorknownexpertiseintheareaofinteresttoyoutobecomethebasisofdatacollection.Yoursampleisagroupofexpertsfromwhomyouseektherequiredinformation.Itislikepurposivesamplingwherethesamplecomprisesexpertsonly.
Explanatoryresearch:Inanexplanatorystudythemainemphasisistoclarifywhyandhowthereisarelationshipbetweentwoaspectsofasituationorphenomenon.
Exploratoryresearch:Thisiswhenastudyisundertakenwiththeobjectiveeithertoexploreanareawherelittleisknownortoinvestigatethepossibilitiesofundertakingaparticularresearchstudy.Whenastudyiscarriedouttodetermineitsfeasibilityitisalsocalledafeasibilityorpilotstudy.
Extraneousvariables:Instudyingcausality,thedependentvariableistheconsequenceofthechangebroughtaboutbytheindependentvariable.Ineverydaylifetherearemanyothervariablesthatcanaffecttherelationshipbetweenindependentanddependentvariables.Thesevariablesarecalledextraneousvariables.
Facevalidity:Whenyoujustifytheinclusionofaquestionoriteminaresearchinstrumentbylinking
itwiththeobjectivesofthestudy,thusprovidingajustificationforitsinclusionintheinstrument,theprocessiscalledfacevalidity.
Feasibilitystudy:Whenthepurposeofastudyistoinvestigatethepossibilityofundertakingitonalargerscaleandtostreamliningmethodsandproceduresforthemainstudy,thestudyiscalledafeasibilitystudy.
Feministresearch:Likeactionresearch,feministresearchismoreaphilosophythandesign.Feministconcernsandtheoryactastheguidingframeworkforthisresearch.Afocusontheviewpointsofwomen,theaimtoreducepowerimbalancebetweenresearcherandrespondents,andattemptstochangesocialinequalitybetweenmenandwomenarethemaincharacteristicsoffeministresearch.
Fishbowldraw:ThisisoneofthemethodsofselectingarandomsampleandisusefulparticularlywhenNisnotverylarge.Itentailswritingeachelementnumberonasmallslipofpaper,foldedandputintoabowl,shufflingthoroughly,andthentakingoneouttilltherequiredsamplesizeisobtained.
Focusgroup:Thefocusgroupisaformofstrategyinqualitativeresearchinwhichattitudes,opinionsorperceptionstowardsanissue,product,serviceorprogrammeareexploredthroughafreeandopendiscussionbetweenmembersofagroupandtheresearcher.Thefocusgroupisafacilitatedgroupdiscussioninwhicharesearcherraisesissuesorasksquestionsthatstimulatediscussionamongmembersofthegroup.Issues,questionsanddifferentperspectivesonthemandanysignificantpointsarisingduringthesediscussionsprovidedatatodrawconclusionsandinferences.Itislikecollectivelyinterviewingagroupofrespondents.
Frameofanalysis:Theproposedplanofthewayyouwanttoanalyseyourdata,howyouaregoingtoanalysethedatatooperationaliseyourmajorconceptsandwhatstatisticalproceduresyouareplanningtouse,allformpartsoftheframeofanalysis.
Frequencydistribution:Thefrequencydistributionisastatisticalprocedureinquantitativeresearchthatcanbeappliedtoanyvariablethatismeasuredonanyoneofthefourmeasurementscales.Itgroupsrespondentsintothesubcategoriesinwhichavariablehasbeenmeasuredorcoded.
Frequencypolygon:Thefrequencypolygonisverysimilartoahistogram.Afrequencypolygonisdrawnbyjoiningthemidpointofeachrectangleataheightcommensuratewiththefrequencyofthatinterval.
Groupinterview:Agroupinterviewisbothamethodofdatacollectionandaqualitativestudydesign.Theinteractionisbetweentheresearcherandthegroupwiththeaimofcollectinginformationfromthegroupcollectivelyratherthanindividuallyfrommembers.
Guttmanscale:TheGuttmanscaleisoneofthethreeattitudinalscalesandisdevisedinsuchawaythatthestatementsoritemsreflectingattitudearearrangedinperfectcumulativeorder.Arrangingstatementsoritemstohaveacumulativerelationbetweenthemisthemostdifficultaspectofconstructingthisscale.
Haloeffect:Whenmakinganobservation,someobserversmaybeinfluencedtorateanindividualononeaspectoftheinteractionbytheways/hewasratedonanother.Thisissimilartosomethingthatcanhappeninteachingwhenateacher’sassessmentoftheperformanceofastudentinonesubjectmay
influencehis/herratingofthatstudent’sperformanceinanother.Thistypeofeffectisknownasthehaloeffect.
Hawthorneeffect:Whenindividualsorgroupsbecomeawarethattheyarebeingobserved,theymaychangetheirbehaviour.Dependinguponthesituation,thischangecouldbepositiveornegative–itmayincreaseordecrease,forexample,theirproductivity–andmayoccurforanumberofreasons.Whenachangeinthebehaviourofpersonsorgroupsisattributedtotheirbeingobserved,itisknownastheHawthorneeffect.
Histogram:Ahistogramisagraphicpresentationofanalyseddatapresentedintheformofaseriesofrectanglesdrawnnexttoeachotherwithoutanyspacebetweenthem,eachrepresentingthefrequencyofacategoryorsubcategory.
Holisticresearchismoreaphilosophythanastudydesign.Thedesignisbaseduponthephilosophythatasamultiplicityoffactorsinteractsinourlives,wecannotunderstandaphenomenonfromoneortwoperspectivesonly.Tounderstandasituationorphenomenonweneedtolookatitinitstotalityorentirety;thatis,holisticallyfromeveryperspective.Aresearchstudydonewiththisphilosophicalperspectiveinmindiscalledholisticresearch.
Hypothesis:Ahypothesisisahunch,assumption,suspicion,assertionoranideaaboutaphenomenon,relationshiporsituation,therealityortruthofwhichyoudonotknowandyousetupyourstudytofindthistruth.Aresearcherreferstotheseassumptions,assertions,statementsorhunchesashypothesesandtheybecomethebasisofanenquiry.Inmoststudiesthehypothesiswillbebasedeitheruponpreviousstudiesoronyourownorsomeoneelse’sobservations.
Hypothesisofassociation:Whenasaresearcheryouhavesufficientknowledgeaboutasituationorphenomenonandareinapositiontostipulatetheextentoftherelationshipbetweentwovariablesandformulateahunchthatreflectsthemagnitudeoftherelationship,suchatypeofhypothesisformulationisknownashypothesisofassociation.
Hypothesisofdifference:Ahypothesisinwhicharesearcherstipulatesthattherewillbeadifferencebutdoesnotspecifyitsmagnitudeiscalledahypothesisofdifference.
Hypothesisofpoint-prevalence:Therearetimeswhenaresearcherhasenoughknowledgeaboutaphenomenonthathe/sheisstudyingandisconfidentaboutspeculatingalmosttheexactprevalenceofthesituationortheoutcomeinquantitativeunits.Thistypeofhypothesisisknownasahypothesisofpoint-prevalence.
Illuminativeevaluation:Theprimaryconcernofilluminativeorholisticevaluationisdescriptionandinterpretationratherthanmeasurementandpredictionofthetotalityofaphenomenon.Itfitswiththesocial–anthropologicalparadigm.Theaimistostudyaprogrammeinallitsaspects:howitoperates,howitisinfluencedbyvariouscontexts,howitisapplied,howthosedirectlyinvolvedviewitsstrengthsandweaknesses,andwhattheexperiencesareofthosewhoareaffectedbyit.Insummary,ittriestoilluminateanarrayofquestionsandissuesrelatingtothecontents,andprocesses,andproceduresthatgivebothdesirableandundesirableresults.
Impactassessmentevaluation:Impactoroutcomeevaluationisoneofthemostwidelypractisedevaluations.Itisusedtoassesswhatchangescanbeattributedtotheintroductionofaparticular
intervention,programmeorpolicy.Itestablishescausalitybetweenaninterventionanditsimpact,andestimatesthemagnitudeofthischange(s).
Independentvariable:Whenexaminingcausalityinastudy,therearefoursetsofvariablesthatcanoperate.Oneofthemisavariablethatisresponsibleforbringingaboutchange.Thisvariablewhichisthecauseofthechangesinaphenomenoniscalledanindependentvariable.Inthestudyofcausality,theindependentvariableisthecausevariablewhichisresponsibleforbringingaboutchangeinaphenomenon.
In-depthinterviewingisanextremelyusefulmethodofdatacollectionthatprovidescompletefreedomintermsofcontentandstructure.Asaresearcheryouarefreetoordertheseinwhateversequenceyouwish,keepinginmindthecontext.Youalsohavecompletefreedomintermsofwhatquestionsyouaskofyourrespondents,thewordingyouuseandthewayyouexplainthemtoyourrespondents.Youusuallyformulatequestionsandraiseissuesonthespurofthemoment,dependinguponwhatoccurstoyouinthecontextofthediscussion.
Indicators:Animage,perceptionorconceptissometimesincapableofdirectmeasurement.Insuchsituationsaconceptis‘measured’throughothermeanswhicharelogically‘reflective’oftheconcept.Theselogicalreflectorsarecalledindicators.
Informedconsentimpliesthatrespondentsaremadeadequatelyandaccuratelyawareofthetypeofinformationyouwantfromthem,whytheinformationisbeingsought,whatpurposeitwillbeputto,howtheyareexpectedtoparticipateinthestudy,andhowitwilldirectlyorindirectlyaffectthem.Itisimportantthattheconsentshouldalsobevoluntaryandwithoutpressureofanykind.Theconsentgivenbyrespondentsafterbeingadequatelyandaccuratelymadeawareoforinformedaboutallaspectsofastudyiscalledinformedconsent.
Interruptedtime-seriesdesign:Inthisdesignyoustudyagroupofpeoplebeforeandaftertheintroductionofanintervention.Itislikethebefore-and-afterdesign,exceptthatyouhavemultipledatacollectionsatdifferenttimeintervalstoconstituteanaggregatedbefore-and-afterpicture.Thedesignisbasedupontheassumptionthatonesetofdataisnotsufficienttoestablish,withareasonabledegreeofcertaintyandaccuracy,thebefore-and-aftersituations.
Intervalscale:Theintervalscaleisoneofthemeasurementscalesinthesocialscienceswherethescaleisdividedintoanumberofintervalsorunits.Anintervalscalehasallthecharacteristicsofanordinalscale.Inaddition,ithasaunitofmeasurementthatenablesindividualsorresponsestobeplacedatequallyspacedintervalsinrelationtothespreadofthescale.Thisscalehasastartingandaterminatingpointandisdividedintoequallyspacedunits/intervals.Thestartingandterminatingpointsandthenumberofunits/intervalsbetweenthemarearbitraryandvaryfromscaletoscaleasitdoesnothaveafixedzeropoint.
Interveningvariableslinktheindependentanddependentvariables.Incertainsituationstherelationshipbetweenanindependentandadependentvariabledoesnoteventuatetilltheinterventionofanothervariable–theinterveningvariable.Thecausevariablewillhavetheassumedeffectonlyinthepresenceofaninterveningvariable.
Intervention–development–evaluationprocess:Thisisacyclicalprocessofcontinuousassessmentofneeds,interventionandevaluation.Youmakeanassessmentoftheneedsofagrouporcommunity,
developinterventionstrategiestomeettheseneeds,implementtheinterventionsandthenevaluatethemformakinginformeddecisionstoincorporatechangestoenhancetheirrelevance,efficiencyandeffectiveness.Reassesstheneedsandfollowthesameprocessforintervention–development–evaluation.
Interviewguide:Alistofissues,topicsordiscussionpointsthatyouwanttocoverinanin-depthinterviewiscalledaninterviewguide.Notethatthesepointsarenotquestions.Itisbasicallyalisttoremindanintervieweroftheareastobecoveredinaninterview.
Interviewschedule:Aninterviewscheduleisawrittenlistofquestions,openendedorclosed,preparedforusebyaninterviewerinaperson-to-personinteraction(thismaybefacetoface,bytelephoneorbyotherelectronicmedia).Notethataninterviewscheduleisaresearchtool/instrumentforcollectingdata,whereasinterviewingisamethodofdatacollection.
Interviewingisoneofthecommonlyusedmethodsofdatacollectioninthesocialsciences.Anyperson-to-personinteraction,eitherfacetofaceorotherwise,betweentwoormoreindividualswithaspecificpurposeinmindiscalledaninterview.Itinvolvesaskingquestionsofrespondentsandrecordingtheiranswers.Interviewingspansawidespectrumintermsofitsstructure.Ontheonehand,itcouldbehighlystructuredand,ontheother,extremelyflexible,andinbetweenitcouldacquireanyform.
Judgementalsampling:Theprimaryconsiderationinthissamplingdesignisyourjudgementastowhocanprovidethebestinformationtoachievetheobjectivesofyourstudy.Youasaresearcheronlygotothosepeoplewhoinyouropinionarelikelytohavetherequiredinformationandarewillingtoshareitwithyou.Thisdesignisalsocalledpurposivesampling.
Leadingquestion:Aleadingquestionisonewhich,byitscontents,structureorwording,leadsarespondenttoanswerinacertaindirection.
Likertscale:TheLikertscale,alsoknownasthesummatedratingscale,isoneoftheattitudinalscalesdesignedtomeasureattitudes.Thisscaleisbasedupontheassumptionthateachstatement/itemonthescalehasequalattitudinal‘value’,‘importance’or‘weight’intermsofreflectingattitudetowardstheissueinquestion.Comparativelyitistheeasiesttoconstruct.
Literaturereview:Thisistheprocessofsearchingtheexistingliteraturerelatingtoyourresearchproblemtodeveloptheoreticalandconceptualframeworksforyourstudyandtointegrateyourresearchfindingswithwhattheliteraturesaysaboutthem.Itplacesyourstudyinperspectivetowhatothershaveinvestigatedabouttheissues.Inadditiontheprocesshelpsyoutoimproveyourmethodology.
Longitudinalstudy:Inlongitudinalstudiesthestudypopulationisvisitedanumberoftimesatregularintervals,usuallyoveralongperiod,tocollecttherequiredinformation.Theseintervalsarenotfixedsotheirlengthmayvaryfromstudytostudy.Intervalsmightbeasshortasaweekorlongerthanayear.Irrespectiveofthesizeoftheinterval,theinformationgatheredeachtimeisidentical.
Matchingisatechniquethatisusedtoformtwogroupsofpatientstosetupanexperiment–controlstudytotesttheeffectivenessofadrug.Fromapoolofpatients,twopatientswithidenticalpredeterminedattributes,characteristicsorconditionsarematchedandthenrandomlyplacedineithertheexperimentalorcontrolgroup.Theprocessiscalledmatching.Thematchingcontinuesfortherest
ofthepool.Thetwogroupsthusformedthroughthematchingprocessaresupposedtobecomparablethusensuringuniformimpactofdifferentsetsofvariablesonthepatients.
Maturationeffect:Ifthestudypopulationisveryyoungandifthereisasignificanttimelapsebetweenthebefore-and-aftersetsofdatacollection,thestudypopulationmaychangesimplybecauseitisgrowingolder.Thisisparticularlytruewhenyouarestudyingyoungchildren.Theeffectofthismaturation,ifitissignificantlycorrelatedwiththedependentvariable,isreflectedatthe‘after’observationandisknownasthematurationeffect.
Maxminconprincipleofvariance:Whenstudyingcausalitybetweentwovariablestherearethreesetsofvariablethatimpactuponthedependentvariable.Sinceyouraimasaresearcheristodeterminethechangethatcanbeattributedtotheindependentvariable,youneedtodesignyourstudytoensurethattheindependentvariablehasthemaximumopportunitytohaveitsfullimpactonthedependentvariable,whiletheeffectsthatareattributedtoextraneousandchancevariablesareminimised.Settingupastudytoachievetheaboveisknownasadheringtothemaxminconprincipleofvariance.
Narratives:Thenarrativetechniqueofgatheringinformationhasevenlessstructurethanthefocusgroup.Narrativeshavealmostnopredeterminedcontentsexceptthattheresearcherseekstohearthepersonalexperienceofapersonwithanincidentorhappeninginhis/herlife.Essentially,thepersontellshis/herstoryaboutanincidentorsituationandyou,astheresearcher,listenpassively,occasionallyencouragingtherespondent.
Nominalscale:Thenominalscaleisoneofthewaysofmeasuringavariableinthesocialsciences.Itenablestheclassificationofindividuals,objectsorresponsesbasedonacommon/sharedpropertyorcharacteristic.Thesepeople,objectsorresponsesaredividedintoanumberofsubgroupsinsuchawaythateachmemberofthesubgrouphasthecommoncharacteristic.
Non-experimentalstudies:Therearetimeswhen,instudyingcausality,aresearcherobservesanoutcomeandwishestoinvestigateitscausation.Fromtheoutcomestheresearcherstartslinkingcauseswiththem.Suchstudiesarecallednon-experimentalstudies.Inanon-experimentalstudyyouneitherintroducenorcontrol/manipulatethecausevariable.Youstartwiththeeffectsandtrytolinkthemwiththecauses.
Non-participantobservation:Whenyou,asaresearcher,donotgetinvolvedintheactivitiesofthegroupbutremainapassiveobserver,watchingandlisteningtoitsactivitiesandinteractionsanddrawingconclusionsfromthem,thisiscallednon-participantobservation.
Non-probabilitysamplingdesignsdonotfollowthetheoryofprobabilityintheselectionofelementsfromthesamplingpopulation.Non-probabilitysamplingdesignsareusedwhenthenumberofelementsinapopulationiseitherunknownorcannotbeindividuallyidentified.Insuchsituationstheselectionofelementsisdependentuponotherconsiderations.Non-probabilitysamplingdesignsarecommonlyusedinbothquantitativeandqualitativeresearch.
Nullhypothesis:Whenyouconstructahypothesisstipulatingthatthereisnodifferencebetweentwosituations,groups,outcomes,ortheprevalenceofaconditionorphenomenon,thisiscalledanullhypothesisandisusuallywrittenasH0.
Objective-orientedevaluation:Thisiswhenanevaluationisdesignedtoascertainwhetherornota
programmeoraserviceisachievingitsobjectivesorgoals.
Observationisoneofthemethodsforcollectingprimarydata.Itisapurposeful,systematicandselectivewayofwatchingandlisteningtoaninteractionorphenomenonasittakesplace.Thoughdominantlyusedinqualitativeresearch,itisalsousedinquantitativeresearch.
Open-endedquestions:Inanopen-endedquestionthepossibleresponsesarenotgiven.Inthecaseofaquestionnaire,arespondentwritesdowntheanswersinhis/herwords,whereasinthecaseofaninterviewscheduletheinvestigatorrecordstheanswerseitherverbatimorinasummarydescribingarespondent’sanswer.
Operationaldefinition:Whenyoudefineconceptsusedbyyoueitherinyourresearchproblemorinthestudypopulationinameasurableform,theyarecalledworkingoroperationaldefinitions.Itisimportantforyoutounderstandthattheworkingdefinitionsthatyoudevelopareonlyforthepurposeofyourstudy.
Oralhistoryismoreamethodofdatacollectionthanastudydesign;however,inqualitativeresearch,ithasbecomeanapproachtostudyahistoricaleventorepisodethattookplaceinthepastorforgaininginformationaboutaculture,customorstorythathasbeenpassedonfromgenerationtogeneration.Itisapictureofsomethinginsomeone’sownwords.Oralhistories,likenarratives,involvetheuseofbothpassiveandactivelistening.Oralhistories,however,aremorecommonlyusedforlearningaboutcultural,socialorhistoricaleventswhereasnarrativesaremoreaboutaperson’sownexperiences.
Ordinalscale:Anordinalscalehasallthepropertiesofanominalscaleplusoneofitsown.Besidescategorisingindividuals,objects,responsesorapropertyintosubgroupsonthebasisofacommoncharacteristic,itranksthesubgroupsinacertainorder.Theyarearrangedineitherascendingordescendingorderaccordingtotheextentthatasubcategoryreflectsthemagnitudeofvariationinthevariable.
Outcomeevaluation:Thefocusofanoutcomeevaluationistofindouttheeffects,impacts,changesoroutcomesthattheprogrammehasproducedinthetargetpopulation.
Panelstudiesareprospectiveinnatureandaredesignedtocollectinformationfromthesamerespondentsoveraperiodoftime.Theselectedgroupofindividualsbecomesapanelthatprovidestherequiredinformation.Inapanelstudytheperiodofdatacollectioncanrangefromonceonlytorepeateddatacollectionsoveralongperiod.
Participantobservationiswhenyou,asaresearcher,participateintheactivitiesofthegroupbeingobservedinthesamemannerasitsmembers,withorwithouttheirknowingthattheyarebeingobserved.Participantobservationisprincipallyusedinqualitativeresearchandisusuallydonebydevelopingacloseinteractionwithmembersofagroupor‘living’inwiththesituationwhichisbeingstudied.
Participatoryresearch:Bothparticipatoryresearchandcollaborativeenquiryarenotstudydesignspersebutsignifyaphilosophicalperspectivethatadvocatesanactiveinvolvementofresearchparticipantsintheresearchprocess.Participatoryresearchisbasedupontheprincipleofminimisingthe‘gap’betweentheresearcherandtheresearchparticipants.Themostimportantfeatureistheinvolvementandparticipationofthecommunityorresearchparticipantsintheresearchprocesstomaketheresearch
findingsmorerelevanttotheirneeds.
Piechart:Thepiechartisanotherwayofrepresentingdatagraphically.Asthereare360degreesinacircle,thefullcirclecanbeusedtorepresent100percentorthetotalpopulation.Thecircleorpieisdividedintosectionsinaccordancewiththemagnitudeofeachsubcategorycomprisingthetotalpopulation.Henceeachsliceofthepieisinproportiontothesizeofeachsubcategoryofafrequencydistribution.
Pilotstudy:SeeFeasibilitystudy
Placeboeffect:Apatient’sbeliefthats/heisreceivingthetreatmentplaysanimportantroleinhis/herrecoveryeventhoughthetreatmentisfakeorineffective.Thechangeoccursbecauseapatientbelievesthats/heisreceivingthetreatment.Thispsychologicaleffectthathelpsapatienttorecoverisknownastheplaceboeffect.
Placebostudy:Astudythatattemptstodeterminetheextentofaplaceboeffectiscalledaplacebostudy.Aplacebostudyisbaseduponacomparativestudydesignthatinvolvestwoormoregroups,dependingonwhetherornotyouwanttohaveacontrolgrouptoisolatetheimpactofextraneousvariablesorothertreatmentmodalitiestodeterminetheirrelativeeffectiveness.
Polytomousvariable:Whenavariablecanbedividedintomorethantwocategories,forexamplereligion(Christian,Muslim,Hindu),politicalparties(Labor,Liberal,Democrat),andattitudes(stronglyfavourable,favourable,uncertain,unfavourable,stronglyunfavourable),itiscalledapolytomousvariable.
Populationmean:Fromwhatyoufindoutfromyoursample(samplestatistics)youmakeanestimateoftheprevalenceofthesecharacteristicsforthetotalstudypopulation.Theestimatesaboutthetotalstudypopulationmadefromsamplestatisticsarecalledpopulationparametersorthepopulationmean.
Predictivevalidityisjudgedbythedegreetowhichaninstrumentcancorrectlyforecastanoutcome:thehigherthecorrectnessintheforecasts,thehigherthepredictivevalidityoftheinstrument.
Pre-test:Inquantitativeresearch,pre-testingisapracticewherebyyoutestsomethingthatyoudevelopedbeforeitsactualusetoascertainthelikelyproblemswithit.Mostly,thepretestisdoneonaresearchinstrumentoronacodebook.Thepre-testofaresearchinstrumententailsacriticalexaminationofeachquestionastoitsclarity,understanding,wordingandmeaningasunderstoodbypotentialrespondentswithaviewtoremovingpossibleproblemswiththequestion.Itensuresthatarespondent’sunderstandingofeachquestionisinaccordancewithyourintentions.Thepre-testofaninstrumentisonlydoneinstructuredstudies.Pre-testingacodebookentailsactuallycodingafewquestionnaires/interviewschedulestoidentifyanyproblemswiththecodebookbeforecodingthedata.
Primarydata:Informationcollectedforthespecificpurposeofastudyeitherbytheresearcherorbysomeoneelseiscalledprimarydata.
Primarysources:Sourcesthatprovideprimarydatasuchasinterviews,observations,andquestionnairesarecalledprimarysources.
Probabilitysampling:Whenselectingasample,ifyouadheretothetheoryofprobability,thatisyou
selectthesampleinsuchawaythateachelementinthestudypopulationhasanequalandindependentchanceofselectioninthesample,theprocessiscalledprobabilitysampling.
Processevaluation:Themainemphasisofprocessevaluationisonevaluatingthemannerinwhichaserviceorprogrammeisbeingdeliveredinordertoidentifywaysofenhancingtheefficiencyofthedeliverysystem.
Programmeplanningevaluation:Beforestartingalarge-scaleprogrammeitisdesirabletoinvestigatetheextentandnatureoftheproblemforwhichtheprogrammeisbeingdeveloped.Whenanevaluationisundertakenwiththepurposeofinvestigatingthenatureandextentoftheproblemitself,itiscalledprogrammeplanningevaluation.
Proportionatestratifiedsampling:Inproportionatestratifiedsampling,thenumberofelementsselectedinthesamplefromeachstratumisinrelationtoitsproportioninthetotalpopulation.Asamplethusselectediscalledaproportionatestratifiedsample.
Prospectivestudiesrefertothelikelyprevalenceofaphenomenon,situation,problem,attitudeoroutcomeinthefuture.Suchstudiesattempttoestablishtheoutcomeofaneventorwhatislikelytohappen.Experimentsareusuallyclassifiedasprospectivestudiesbecausetheresearchermustwaitforaninterventiontoregisteritseffectonthestudypopulation.
Pureresearchisconcernedwiththedevelopment,examination,verificationandrefinementofresearchmethods,procedures,techniquesandtoolsthatformthebodyofresearchmethodology.
Purposivesampling:SeeJudgementalsampling
Qualitativeresearch:Inthesocialsciencestherearetwobroadapproachestoenquiry:qualitativeandquantitativeorunstructuredandstructuredapproaches.Qualitativeresearchisbaseduponthephilosophyofempiricism,followsanunstructured,flexibleandopenapproachtoenquiry,aimstodescribethanmeasure,believesinin-depthunderstandingandsmallsamples,andexploresperceptionsandfeelingsthanfactsandfigures.
Quantitativeresearchisasecondapproachtoenquiryinthesocialsciencesthatisrootedinrationalism,followsastructured,rigid,predeterminedmethodology,believesinhavinganarrowfocus,emphasisesgreatersamplesize,aimstoquantifythevariationinaphenomenon,andtriestomakegeneralisationstothetotalpopulation.
Quasi-experiments:Studieswhichhavetheattributesofbothexperimentalandnon-experimentalstudiesarecalledquasi-orsemi-experiments.Apartofthestudycouldbeexperimentalandtheothernon-experimental.
Questionnaire:Aquestionnaireisawrittenlistofquestions,theanswerstowhicharerecordedbyrespondents.Inaquestionnairerespondentsreadthequestions,interpretwhatisexpectedandthenwritedowntheanswers.Theonlydifferencebetweenaninterviewscheduleandaquestionnaireisthatintheformeritistheinterviewerwhoasksthequestions(and,ifnecessary,explainsthem)andrecordstherespondent’srepliesonaninterviewschedule,whileinthelatterrepliesarerecordedbytherespondentsthemselves.
Quotasampling:Themainconsiderationdirectingquotasamplingistheresearcher’seaseofaccesstothesamplepopulation.Inadditiontoconvenience,aresearcherisguidedbysomevisiblecharacteristicofinterest,suchasgenderorrace,ofthestudypopulation.Thesampleisselectedfromalocationconvenienttoyouasaresearcher,andwheneverapersonwiththisvisiblerelevantcharacteristicisseen,thatpersonisaskedtoparticipateinthestudy.Theprocesscontinuesuntilyouhavebeenabletocontacttherequirednumberofrespondents(quota).
Randomdesign:Inarandomdesign,thestudypopulationgroupsaswellastheexperimentaltreatmentsarenotpredeterminedbutrandomlyassignedtobecomecontrolorexperimentalgroups.Randomassignmentinexperimentsmeansthatanyindividualorunitofthestudypopulationhasanequalandindependentchanceofbecomingapartoftheexperimentalorcontrolgroupor,inthecaseofmultipletreatmentmodalities,anytreatmenthasanequalandindependentchanceofbeingassignedtoanyofthepopulationgroups.Itisimportanttonotethattheconceptofrandomisationcanbeappliedtoanyoftheexperimentaldesigns.
Randomsampling:Foradesigntobecalledrandomorprobabilitysampling,itisimperativethateachelementinthestudypopulationhasanequalandindependentchanceofselectioninthesample.Equalimpliesthattheprobabilityofselectionofeachelementinthestudypopulationisthesame.Theconceptofindependencemeansthatthechoiceofoneelementisnotdependentuponthechoiceofanotherelementinthesampling.
Randomvariable:Whencollectinginformationfromrespondents,therearetimeswhenthemoodofarespondentorthewordingofaquestioncanaffectthewayarespondentreplies.Thereisnosystematicpatternintermsofthischange.Suchshiftsinresponsesaresaidtobecausedbyrandomorchancevariables.
Randomisation:Inexperimentalandcomparativestudies,youoftenneedtostudytwoormoregroupsofpeople.Informingthesegroupsitisimportantthattheyarecomparablewithrespecttothedependentvariableandothervariablesthataffectitsothattheeffectsofindependentandextraneousvariablesareuniformacrossgroups.Randomisationisaprocessthatensuresthateachandeverypersoninagroupisgivenanequalandindependentchanceofbeinginanyofthegroups,therebymakinggroupscomparable.
Ratioscale:Aratioscalehasallthepropertiesofnominal,ordinalandintervalscalesplusitsownproperty;thezeropointofaratioscaleisfixed,whichmeansithasafixedstartingpoint.Therefore,itisanabsolutescale.Asthedifferencebetweentheintervalsisalwaysmeasuredfromazeropoint,arithmeticaloperationscanbeperformedonthescores.
Reactiveeffect:Sometimesthewayaquestioniswordedinformsrespondentsoftheexistenceorprevalenceofsomethingthatthestudyistryingtofindoutaboutasanoutcomeofanintervention.Thiseffectisknownasreactiveeffectoftheinstrument
Recallerror:Errorthatcanbeintroducedinaresponsebecauseofarespondent’sinabilitytorecallcorrectlyitsvariousaspectswhenreplying.
Regressioneffect:Sometimespeoplewhoplacethemselvesontheextremepositionsofameasurementscaleatthepre-teststagemay,foranumberofreasons,shifttowardsthemeanatthepost-teststage.Theymightfeelthattheyhavebeentoonegativeortoopositiveatthepre-teststage.Therefore,the
mereexpressionoftheattitudeinresponsetoaquestionnaireorinterviewhascausedthemtothinkaboutandaltertheirattitudetowardsthemeanatthetimeofthepost-test.Thistypeofeffectisknownastheregressioneffect.
Reflectivejournallog:Basicallythisisamethodofdatacollectioninqualitativeresearchthatentailskeepingalogofyourthoughtsasaresearcherwheneveryounoticeanything,talktosomeone,participateinanactivityorobservesomethingthathelpsyouunderstandoraddtowhateveryouaretryingtofindoutabout.Thislogbecomesthebasisofyourresearchfindings.
Reflexivecontroldesign:Inexperimentalstudies,toovercometheproblemofcomparabilityindifferentgroups,sometimesresearchersstudyonlyonepopulationandtreatdatacollectedduringthenon-interventionperiodasrepresentingacontrolgroup,andinformationcollectedaftertheintroductionoftheinterventionasifitpertainedtoanexperimentalgroup.Itistheperiodsofnon-interventionandinterventionthatconstitutecontrolandexperimentalgroups.
Reliabilityistheabilityofaresearchinstrumenttoprovidesimilarresultswhenusedrepeatedlyundersimilarconditions.Reliabilityindicatesaccuracy,stabilityandpredictabilityofaresearchinstrument:thehigherthereliability,thehighertheaccuracy;orthehighertheaccuracyofaninstrument,thehigheritsreliability.
Replicatedcross-sectionaldesign:Thisstudydesignisbasedupontheassumptionthatparticipantsatdifferentstagesofaprogrammearesimilarintermsoftheirsocioeconomic–demographiccharacteristicsandtheproblemforwhichtheyareseekingintervention.Assessmentoftheeffectivenessofaninterventionisdonebytakingasampleofclientswhoareatdifferentstagesoftheintervention.Thedifferenceinthedependentvariableamongclientsattheintakeandterminationstageisconsideredtobetheimpactoftheintervention.
Researchisoneofthewaysoffindinganswerstoyourprofessionalandpracticequestions.However,itischaracterisedbytheuseoftestedproceduresandmethodsandanunbiasedandobjectiveattitudeintheprocessofexploration.
Researchdesign:Aresearchdesignisaproceduralplanthatisadoptedbytheresearchertoanswerquestionsvalidly,objectively,accuratelyandeconomically.Aresearchdesignthereforeanswersquestionsthatwoulddeterminethepathyouareproposingtotakeforyourresearchjourney.Througharesearchdesignyoudecideforyourselfandcommunicatetoothersyourdecisionsregardingwhatstudydesignyouproposetouse,howyouaregoingtocollectinformationfromyourrespondents,howyouaregoingtoselectyourrespondents,howtheinformationyouaregoingtocollectistobeanalysedandhowyouaregoingtocommunicateyourfindings.
Researchobjectivesarespecificstatementsofgoalsthatyousetouttobeachievedattheendofyourresearchjourney.
Researchproblem:Anyissue,problemorquestionthatbecomesthebasisofyourenquiryiscalledaresearchproblem.Itiswhatyouwanttofindoutaboutduringyourresearchendeavour.
Researchquestions:Questionsthatyouwouldliketofindanswerstothroughyourresearch,like‘WhatdoesitmeantohaveachildwithADHDinafamily?’or‘Whatistheimpactofimmigrationonfamilyroles?’Researchquestionsbecomethebasisofresearchobjectives.Themaindifferencebetween
researchquestionsandresearchobjectivesisthewaytheyareworded.Researchquestionstaketheformofquestionswhereasresearchobjectivesarestatementsofachievementsexpressedusingaction-orientedwords.
Retrospectivestudy:Aretrospectivestudyinvestigatesaphenomenon,situation,problemorissuethathashappenedinthepast.Suchstudiesareusuallyconductedeitheronthebasisofthedataavailableforthatperiodoronthebasisofrespondents’recallofthesituation.
Retrospective–prospectivestudy:Aretrospective–prospectivestudyfocusesonpasttrendsinaphenomenonandstudiesitintothefuture.Astudywhereyoumeasuretheimpactofaninterventionwithouthavingacontrolgroupby‘constructing’apreviousbaselinefromeitherrespondents’recallorsecondarysources,thenintroducingtheinterventiontostudyitseffect,isconsideredaretrospective–prospectivestudy.Infact,mostbefore-and-afterstudies,ifcarriedoutwithouthavingacontrol–wherethebaselineisconstructedfromthesamepopulationbeforeintroducingtheintervention–willbeclassifiedasretrospective-prospectivestudies.
Rowpercentagesarecalculatedfromthetotalofallthesubcategoriesofonevariablethataredisplayedalongarowindifferentcolumns.
Sample:Asampleisasubgroupofthepopulationwhichisthefocusofyourresearchenquiryandisselectedinsuchawaythatitrepresentsthestudypopulation.Asampleiscomposedofafewindividualsfromwhomyoucollecttherequiredinformation.Itisdonetosavetime,moneyandotherresources.
Samplesize:Thenumberofindividualsfromwhomyouobtaintherequiredinformationiscalledthesamplesizeandisusuallydenotedbythelettern.
Samplestatistics:Findingsbasedontheinformationobtainedfromyourrespondents(sample)arecalledsamplestatistics.
Samplingistheprocessofselectingafewrespondents(asample)fromabiggergroup(thesamplingpopulation)tobecomethebasisforestimatingtheprevalenceofinformationofinteresttoyou.
Samplingdesign:Thewayyouselecttherequiredsamplingunitsfromasamplingpopulationforidentifyingyoursampleiscalledthesamplingdesignorsamplingstrategy.Therearemanysamplingstrategiesinbothquantitativeandqualitativeresearch.
Samplingelement:Anythingthatbecomesthebasisofselectingyoursamplesuchasanindividual,family,household,membersofanorganisation,residentsofanarea,iscalledasamplingunitorelement.
Samplingerror:Thedifferenceinthefindings(samplestatistics)thatisduetotheselectionofelementsinthesampleisknownassamplingerror.
Samplingframe:Whenyouareinapositiontoidentifyallelementsofastudypopulation,thelistofalltheelementsiscalledasamplingframe.
Samplingpopulation:Thebiggergroup,suchasfamilieslivinginanarea,clientsofanagency,
residentsofacommunity,membersofagroup,peoplebelongingtoanorganisationaboutwhomyouwanttofindoutaboutthroughyourresearchendeavour,iscalledthesamplingpopulationorstudypopulation.
Samplingstrategy:SeeSamplingdesign
Samplingunit:SeeSamplingelement
Samplingwithreplacement:Whenyouselectasampleinsuchawaythateachselectedelementinthesampleisreplacedbackintothesamplingpopulationbeforeselectingthenext,thisiscalledsamplingwithreplacement.Theoretically,thisisdonetoprovideanequalchanceofselectiontoeachelementsoastoadheretothetheoryofprobabilitytoensurerandomisationofthesample.Incaseanelementisselectedagain,itisdiscardedandthenextoneisselected.Ifthesamplingpopulationisfairlylarge,theprobabilityofselectingthesameelementtwiceisfairlyremote.
Samplingwithoutreplacement:Whenyouselectasampleinsuchawaythatanelement,onceselectedtobecomeapartofyoursample,isnotreplacedbackintothestudypopulation,thisiscalledsamplingwithoutreplacement.
Saturationpoint:Theconceptofsaturationpointreferstothestageindatacollectionwhereyou,asaresearcher,arediscoveringnoorverylittlenewinformationfromyourrespondents.Inqualitativeresearchthisisconsideredanindicationoftheadequacyofthesamplesize.
Scale:Thisisamethodofmeasurementand/orclassificationofrespondentsonthebasisoftheirresponsestoquestionsyouaskoftheminastudy.Ascalecouldbecontinuousorcategorical.Ithelpsyoutoclassifyastudypopulationinsubgroupsorasaspreadthatisreflectiveonthescale.
Scattergram:Whenyouwanttoshowgraphicallyhowonevariablechangesinrelationtoachangeintheother,ascattergramisextremelyeffective.Forascattergram,boththevariablesmustbemeasuredeitheronanintervalorratioscaleandthedataonboththevariablesneedstobeavailableinabsolutevaluesforeachobservation.Dataforbothvariablesistakeninpairsanddisplayedasdotsinrelationtotheirvaluesonbothaxes.Theresultinggraphisknownasascattergram.
Secondarydata:Sometimestheinformationrequiredisalreadyavailableinothersourcessuchasjournals,previousreports,censusesandyouextractthatinformationforthespecificpurposeofyourstudy.Thistypeofdatawhichalreadyexistsbutyouextractforthepurposeofyourstudyiscalledsecondarydata.
Secondarysources:Sourcesthatprovidesecondarydataarecalledsecondarysources.Sourcessuchasbooks,journals,previousresearchstudies,recordsofanagency,clientorpatientinformationalreadycollectedandroutineservicedeliveryrecordsallformsecondarysources.
Semi-experimentalstudies:Asemi-experimentaldesignhasthepropertiesofbothexperimentalandnon-experimentalstudies;partofthestudymaybenon-experimentalandtheotherpartexperimental.
Simplerandomsampling:Thisisthemostcommonlyusedmethodofselectingarandomsample.Itisaprocessofselectingtherequiredsamplesizefromthesamplingpopulation,providingeachelementwithanequalandindependentchanceofselectionbyanymethoddesignedtoselectarandomsample.
Snowballsamplingisaprocessofselectingasampleusingnetworks.Tostartwith,afewindividualsinagroupororganisationareselectedusingpurposive,randomornetworksamplingtocollecttherequiredinformationfromthem.Theyarethenaskedtoidentifyotherpeopleinthegroupororganisationwhocouldbecontactedtoobtainthesameinformation.Thepeopleselectedbythembecomeapartofthesample.Theprocesscontinuestillyoureachthesaturationpointintermsofinformationbeingcollected.
Stackedbarchart:Astackedbarchartissimilartoabarchartexceptthatintheformereachbarshowsinformationabouttwoormorevariablesstackedontoeachothervertically.Thesectionsofabarshowtheproportionofthevariablestheyrepresentinrelationtooneanother.Thestackedbarscanbedrawnonlyforcategoricaldata.
Stakeholdersinresearch:Thosepeopleorgroupswhoarelikelytobeaffectedbyaresearchactivityoritsfindings.Inresearchtherearethreestakeholders:theresearchparticipants,theresearcherandthefundingbody.
Stem-and-leafdisplay:Thestem-and-leafdisplayisaneffective,quickandsimplewayofdisplayingafrequencydistribution.Thestemandleafforafrequencydistributionrunningintotwodigitsisplottedbydisplayingdigits0to9ontheleftofthey-axis,representingthetensofafrequency.Thefiguresrepresentingtheunitsofafrequency(i.e.theright-handfigureofatwo-digitfrequency)aredisplayedontherightofthey-axis.
Stratifiedrandomsamplingisoneoftheprobabilitysamplingdesignsinwhichthetotalstudypopulationisfirstclassifiedintodifferentsubgroupsbaseduponacharacteristicthatmakeseachsubgroupmorehomogeneousintermsoftheclassificatoryvariable.Thesampleisthenselectedfromeachsubgroupeitherbyselectinganequalnumberofelementsfromeachsubgrouporselectingelementsfromeachsubgroupequaltoitsproportioninthetotalpopulation.
Stubisapartofthetablestructure.Itisthesubcategoriesofavariable,listedalongthey-axis(theleft-handcolumnofthetable).Thestub,usuallythefirstcolumnontheleft,liststheitemsaboutwhichinformationisprovidedinthehorizontalrowstotheright.Itistheverticallistingofcategoriesorindividualsaboutwhichinformationisgiveninthecolumnsofthetable.
Studydesign:Thetermstudydesignisusedtodescribethetypeofdesignyouaregoingtoadopttoundertakeyourstudy;thatis,ifitisgoingtobeexperimental,correlational,descriptiveorbeforeandafter.Eachstudydesignhasaspecificformatandattributes.
Studypopulation:Everystudyinthesocialscienceshastwoaspects:studypopulationandstudyarea(subjectarea).PeoplewhoyouwanttofindoutaboutarecollectivelyknownasthestudypopulationorsimplypopulationandareusuallydenotedbytheletterN.Itcouldbeagroupofpeoplelivinginanarea,employeesofanorganisation,acommunity,agroupofpeoplewithspecialissues,etc.Thepeoplefromwhomyougatherinformation,knownasthesamplen,areselectedfromthestudypopulation.
Subjectarea:Anyacademicorpracticefieldinwhichyouareconductingyourstudyiscalledthesubjectorstudyarea.Itcouldbehealthorotherneedsofacommunity,attitudesofpeopletowardsanissue,occupationalmobilityinacommunity,copingstrategies,depression,domesticviolence,etc.
Subjectivityisanintegralpartofyourwayofthinkingthatis‘conditioned’byyoureducational
background,discipline,philosophy,experienceandskills.Biasisadeliberateattempttochangeorhighlightsomethingwhichinrealityisnottherebutyoudoitbecauseofyourvestedinterest.Subjectivityisnotdeliberate,itisthewayyouunderstandorinterpretsomething.
Summatedratingscale:SeeLikertscale
Systematicsamplingisawayofselectingasamplewherethesamplingframe,dependinguponthesamplesize,isfirstdividedintoanumberofsegmentscalledintervals.Then,fromthefirstinterval,usingtheSRStechnique,oneelementisselected.Theselectionofsubsequentelementsfromotherintervalsisdependentupontheorderoftheelementselectedinthefirstinterval.Ifinthefirstintervalitisthefifthelement,thefifthelementofeachsubsequentintervalwillbechosen.
Tableofrandomnumbers:Mostbooksonresearchmethodologyandstatisticshavetablesthatcontainrandomlygeneratednumbers.Thereisaspecificwayofselectingarandomsampleusingthesetables.
Tablesofferausefulwayofpresentinganalyseddatainasmallspacethatbringsclaritytothetextandservesasaquickpointofreference.Therearedifferenttypesoftableshousingdatapertainingtoone,twoormorevariables.
Thematicwriting:Astyleofwritingwhichiswrittenaroundmainthemes.
Theoreticalframework:Asyoustartreadingtheliterature,youwillsoondiscoverthattheproblemyouwishtoinvestigatehasitsrootsinanumberoftheoriesthathavebeendevelopedfromdifferentperspectives.Theinformationobtainedfromdifferentsourcesneedstobesortedunderthemainthemesandtheories,highlightingagreementsanddisagreementsamongtheauthors.Thisprocessofstructuringa‘network’ofthesetheoriesthatdirectlyorindirectlyhasabearingonyourresearchtopiciscalledthetheoreticalframework.
Theoryofcausality:Thetheoryofcausalityadvocatesthatinstudyingcauseandeffecttherearethreesetsofvariablesthatareresponsibleforthechange.Theseare:causeorindependentvariable,extraneousvariablesandchangevariables.Itisthecombinationofallthreethatproduceschangeinaphenomenon.
Thurstonescale:TheThurstonescaleisoneofthescalesdesignedtomeasureattitudesinthesocialsciences.Attitudethroughthisscaleismeasuredbymeansofasetofstatements,the‘attitudinalvalue’ofwhichhasbeendeterminedbyagroupofjudges.Arespondent’sagreementwiththestatementassignsascoreequivalenttothe‘attitudinalvalue’ofthestatement.Thetotalscoreofallstatementsistheattitudinalscoreforarespondent.
Transferability:Theconceptoftransferabilityreferstothedegreetowhichtheresultsofqualitativeresearchcanbegeneralisedortransferredtoothercontextsorsettings.
Trendcurve:Asetofdatameasuredonanintervaloraratioscalecanbedisplayedusingalinediagramortrendcurve.Atrendlinecanbedrawnfordatapertainingtobothaspecifictimeandaperiod.Ifitrelatestoaperiod,themidpointofeachintervalataheightcommensuratewitheachfrequencyismarkedasadot.Thesedotsarethenconnectedwithstraightlinestoexaminetrendsinaphenomenon.Ifthedatapertainstoanexacttime,apointisplottedataheightcommensuratewiththefrequencyandalineisthendrawntoexaminethetrend.
Trendstudies:Thesestudiesinvolveselectinganumberofdataobservationpointsinthepast,togetherwithapictureofthepresentorimmediatepastwithrespecttothephenomenonunderstudy,andthenmakingcertainassumptionsastothelikelyfuturetrends.Inawayyouarecompilingacross-sectionalpictureofthetrendsbeingobservedatdifferentpointsintimeoverthepast,presentandfuture.Fromthesecross-sectionalobservationsyoudrawconclusionsaboutthepatternofchange.
TypeIerror:Intestingahypothesis,manyreasonsyoumaysometimescommitamistakeanddrawthewrongconclusionwithrespecttothevalidityofyourhypothesis.Ifyourejectanullhypothesiswhenitistrueandyoushouldnothaverejectedit,thisiscalledaTypeIerror.
TypeIIError:Intestingahypothesis,formanyreasonsyoumaysometimescommitamistakeanddrawthewrongconclusionintermsofthevalidityofyourhypothesis.IfyouacceptanullhypothesiswhenitisfalseandyoushouldnothaveaccepteditthisiscalledaTypeIIerror.
Unethical:Anyprofessionalactivitythatisnotinaccordancewiththeacceptedcodeofconductforthatprofessionisconsideredunethical.
Validity:Theconceptofvaliditycanbeappliedtoeveryaspectoftheresearchprocess.Initssimplestform,validityreferstotheappropriatenessofeachstepinfindingoutwhatyousetoutto.However,theconceptofvalidityismoreassociatedwithmeasurementprocedures.Intermsofthemeasurementprocedure,validityistheabilityofaninstrumenttomeasurewhatitisdesignedtomeasure.
Variable:Animage,perceptionorconceptthatiscapableofmeasurement–hencecapableoftakingondifferentvalues–iscalledavariable.Inotherwords,aconceptthatcanbemeasurediscalledavariable.Avariableisapropertythattakesondifferentvalues.Itisarationalunitofmeasurementthatcanassumeanyoneofanumberofdesignatedsetsofvalues.
Workingdefinition:SeeOperationaldefinition
Bibliography
Ackroyd,Stephen&JohnHughes,1992,DataCollectioninContext,NewYork,Longman.Alkin,MarvinC.&LewisC.Solomon(eds),1983,TheCostsofEvaluation,BeverlyHills,CA,Sage.Alwin,DuaneF.(eds),1978,SurveyDesignandAnalysis:CurrentIssues,BeverlyHills,CA,Sage.Babbie,Earl,1989,SurveyResearchMethods(2ndedn),Belmont,CA,Wadsworth.Babbie,Earl,2007,ThePracticeofSocialResearch(11thedn),Belmont,CA,Wadsworth.Bailey,KennethD.,1978,MethodsofSocialResearch(3rdedn),NewYork,FreePress.Barton, Ruth, 1988, Understanding Social Statistics: An Introduction to Descriptive Statistics, PerthEditorial/Publication Unit, Division of Arts, Education and Social Sciences, Curtin University ofTechnology.
Bernard,H.Russell,1994,ResearchMethodsinAnthropology:QualitativeandQuantitativeApproaches(2ndedn),ThousandOaks,CA,Sage.
Bernard,H.Russell,2000,SocialResearchMethods:QualitativeandQuantitativeApproaches,ThousandOaks,CA,Sage.
Bilson,Andy(ed.),2005,Evidence-basedPracticeinSocialWork,London,Whiting&Birch.Black,JamesA.&DeanJ.Champion,1976,MethodsandIssuesinSocialResearch,NewYork,Wiley.Blaikie,Norman,2007,ApproachestoSocialEnquiry(2ndedn),Cambridge,PolityPress.Bogdan, Robert & Sari Knopp Biklen, 1992,Qualitative Research for Education: An Introduction toTheoryandMethods(2ndedn),Boston,Allyn&Bacon.
Bradburn, M. Norman & Seymour Sudman, 1979, Improving Interview Method and QuestionnaireDesign,London,Jossey-Bass.
Brinberg,David&JosephE.McGreth,1985,ValidityandtheResearchProcess,BeverlyHills,CA,Sage.Bulmer,Martin(eds),1977,SociologicalResearchMethods:AnIntroduction,London,Macmillan.Burns,RobertB.,1997,IntroductiontoResearchMethods(2ndedn),Melbourne,LongmanCheshire.Butcher,Judith,1981,Copy-Editing:TheCambridgeHandbookforEditors,AuthorsandPublishers(2ndedn),Cambridge,CambridgeUniversityPress.
Carr,Wilfred&StevenKemmis,1986,BecomingCritical:Education,KnowledgeandActionResearch,Melbourne,DeakinUniversity.
Cherry,Nita,2002,ActionResearch:APathwaytoAction,KnowledgeandLearning,Melbourne,RMITPublishing.
TheChicagoManualofStyle(14thedn),1993,ChicagoandLondon,UniversityofChicagoPress.Cohen, Morris R. & Ernest Nagel, 1966, An Introduction to Logic and Scientific Methods, London,Routledge&KeganPaul.
CollinsDictionaryoftheEnglishLanguage,1979,Sydney,Collins.CommonwealthofAustralia,2002,StyleManual(6thedn),Milton,WileyAustralia.Cozby,C.Paul,1985,MethodsinBehavioralResearch,London,Mayfield.Crano,WilliamD.&MarilynneB.Brewer,2002,PrincipleandMethodsofSocialResearch (2ndedn),London,LawrenceErlbaum.
Creswell, JohnW., 2003,Research Design: Qualitative, Quantitative andMixedMethods Approaches(2ndedn),ThousandOaks,CA,Sage.
Creswell, JohnW., 2007,Qualitative Inquiry andResearchDesign:Choosing amongFiveApproaches(2ndedn),ThousandOaks,CA,Sage.
Crotty,Michael,1998,TheFoundationsofSocialResearch,StLeonards,NSW,Allen&Unwin.Cunningham,J.Barton,1993,ActionResearchandOrganizationalDevelopment,London,Praeger.deVaus,David,2002,SurveysinSocialResearch(5thedn),StLeonards,NSW,Allen&Unwin.Denzin,NormanK.&YvonnaS.Lincoln(eds),1994,HandbookofQualitativeResearch,ThousandOaks,CA,Sage.
Denzin,NormanK.&YvonnaS.Lincoln(eds),1998,CollectingandInterpretingQualitativeMaterials,ThousandOaks,CA,Sage.
Denzin,NormanK.&YvonnaS.Lincoln(eds),1998,StrategiesofQualitativeInquiry,ThousandOaks,CA,Sage.
Denzin,NormanK.&YvonnaS.Lincoln(eds),2005,TheSageHandbookofQualitativeResearch(3rdedn),ThousandOaks,CA,Sage.
Denzin,NormanK.&YvonnaS.Lincoln,2008,TheLandscapeofQualitativeResearch (3rdedn),LosAngeles,Sage.
Dixon,Beverly&GaryBouma,1984,TheResearchProcess,Melbourne,OxfordUniversityPress.Duncan,OtisDudley, 1984,Notes onSocialMeasurement:Historical andCritical,NewYork,RussellSageFoundation.
Elliott,Jane,2005,UsingNarrativeinSocialResearch,London,Sage.Foreman,E.K.,1991,SurveySamplingPrinciples,NewYork,MarcelDekker.Festinger,Leon&DanielKatz(eds),1966,ResearchMethodsinBehavioralSciences,NewYork,Holt,RinehartandWinston.
Gilbert,Nigel(ed.),2008,ResearchingSocialLife(3rdedn),London,Sage.Gray,Mel,DebbiePlath&StephenWebb,2009,Evidence-basedSocialWork,London,Routledge.Grinnell,RichardJr(ed.),1981,SocialWorkResearchandEvaluation,Itasca,IL,F.E.Peacock.Grinnell,RichardJr(ed.),1988,SocialWorkResearchandEvaluation(3rdedn),Itasca,IL,F.E.Peacock.Grinnell,RichardJr(eds),1993,SocialWorkResearchandEvaluation(4thedn),Itasca,IL,F.E.Peacock.Hakim, Catherine, 1987, Research Design: Strategies and Choices in the Design of Social Research,London,Allen&Unwin.
Hall, Irene & David Hall, 2004, Evaluation and Social Research: Introducing Small Scale Practice,Basingstoke,PalgraveMacmillan.
Hawe, Penelope, Deirdre Degeling & Jane Hall, 1992, Evaluating Health Promotion, Sydney,Maclennan+Petty.
Hessler,RichardM.,1992,SocialResearchMethods,NewYork,WestPublishing.Kaplan, David (eds), 2004,The Sage Handbook of QuantitativeMethodology for the Social Sciences,ThousandOaks,CA,Sage.
Kazdin,Alan,1982,Single-caseResearchDesigns,NewYork,OxfordUniversityPress.Kerlinger,FredN.,1973,FoundationsofBehavioralResearch(2ndedn),NewYork,Holt,RinehartandWinston.
Kerlinger, Fred N., 1979, Behavioral Research: A Conceptual Approach, Sydney, Holt, Rinehart andWinston.
Kerlinger,FredN.,1986,FoundationsofBehavioralResearch (3rdedn),NewYork,Holt,RinehartandWinston.
Kirk,Jerome&MarcL.Miller,1986,ReliabilityandValidityinQualitativeResearch,NewburyPark,CA,Sage.
Krueger,Richard&MaryAnneCasey,2000,FocusGroups:APracticalGuideforAppliedResearch(4thedn),ThousandOaks,CA,Sage.
Kvale,Steinar&SvendBrinkmann,2009,Interviews(2ndedn),ThousandOaks,CA,Sage.Leary, Mark R., 1995, Introduction to Behavioral Research Methods (2nd edn), Pacific Grove, CA,
Brooks/Cole.Longyear,Marie(eds),1983,TheMcGraw-HillStyleManual,AConciseGuideforWritersandEditors,NewYork,McGraw-Hill.
Lundberg,GeorgeA., 1942, Social Research: A Study inMethods of GatheringData (2nd edn), NewYork,Longmans,Green.
Marshall,Catherine&GretchenB.Rossman,2006,DesigningQualitativeResearch,ThousandOaks,CA,Sage.
Martin,DavidW.,1985,DoingPsychologicalExperiments(2ndedn),Monterey,CA,Brooks/Cole.May,Tim,1997,SocialResearch:Issues,MethodsandProcess(2ndedn),Buckingham,OpenUniversityPress.
Minium, EdwardW., 1978, Statistical Reasoning in Psychology and Education (2nd edn), New York,Wiley.
Monette,DuaneR.,ThomasJ.Sullivan&CornellR.DeJong,1986,AppliedSocialResearch:ToolsfortheHumanServices,ForthWorth,TX,Holt,Rinehart,andWinston.
Moser,ClausA.&GrahamKalton,1989,SurveyMethods inSocial Investigation (2ndedn),Aldershot,Gower.
Nachmias,David&ClaraNachmias,1987,ResearchMethodsinSocialSciences,NewYork,StMartin’sPress.
Padgett,DeborahK.,2008,QualitativeMethodsinSocialWorkResearch(2ndedn),ThousandOaks,CA,Sage.
Patton,MichaelQuinn,1990,QualitativeEvaluationandResearchMethods,NewburyPark,CA,Sage.Plano Clark, Vicki L. & John W. Creswell, 2009, Understanding Research: A Consumer’s Guide,Melbourne,Pearson(Merrill).
Poincaré,1952,ScienceandHypothesis,NewYork,Dover.Powers,GeraldT.,ThomasM.Meenaghan&BeverleyG.Twoomey,1985,PracticeFocusedResearch:IntegratingHumanPracticeandResearch,EnglewoodCliffs,NJ,PrenticeHall.
Ritchie,DonaldA.,2003,DoingOralHistory:APracticalGuide,Oxford,OxfordUniversityPress.Robson,Colin,2002,RealWorldResearch(2ndedn),Oxford,Blackwell.Rohlf,F.James&RobertR.Sokal,1969,StatisticalTables,SanFrancisco,CA,W.H.Freeman.Rossi, PeterH.&HowardE.Freeman, 1993,Evaluation:ASystematicApproach,NewburyPark,CA,Sage.
Rossi,PeterH.,HowardE.Freeman&MarkW.Lipsey,1999,Evaluation:ASystematicApproach (6thedn),ThousandOaks,CA,Sage.
Rutman,Leonard(eds),1977,EvaluationResearchMethods:ABasicGuide,BeverlyHills,CA,Sage.Rutman,Leonard,1980,PlanningUsefulEvaluations:AvailabilityAssessment,BeverlyHills,CA,Sage.Rutman, Leonard, 1984,Evaluation Research Methods: A Basic Guide (2nd edn), Beverly Hills, CA,Sage.
Sandefur,GaryD.,HowardE.Freeman&PeterH.Rossi,1986,WorkbookforEvaluation:ASystematicApproach(3rdedn),BeverlyHills,CA,Sage.
Sarantakos,S.,2005,SocialResearch(3rdedn),NewYork,PalgraveMacmillan.Schinke,StevenP.&LewayneGilchrist,1993,‘Ethicsinresearch’,inR.M.Grinnell(eds),SocialWork,ResearchandEvaluation(4thedn),Itasca,IL,F.E.Peacock.
Selltiz,Jahoda,MortonDeutsch&StuartCook,1962,ResearchMethods inSocialRelations (rev.edn),NewYork,Holt,RinehartandWinston.
Siegel,Sidney,1956,NonparametricStatisticsfortheBehavioralSciences,Sydney,McGraw-Hill.Silverman,David,2004,QualitativeResearch:Theory,MethodsandPractice(2ndedn),London,Sage.Silverman,David,2005,DoingQualitativeResearch(2ndedn),London,Sage.
Simon,JulianL.,1969,BasicResearchMethodsinSocialSciences:TheArtofEmpirical Investigation,NewYork,RandomHouse.
Smith, Herman W., 1991, Strategies of Social Research (3rd edn), Orlando, FL, Holt, Rinehart andWinston.
Somekh,BridgetandCathyLewin (eds),2005,ResearchMethods in theSocialSciences,LosAngeles,CA,Sage.
Stake,RobertE.,1995,TheArtofCaseStudyResearch,ThousandOaks,CA,Sage.Stevens, Stanley Smith, 1951, ‘Mathematics,measurement, and psychophysics’, in S. S. Stevens (ed.),HandbookofExperimentalPsychology,NewYork,Wiley.
Stufflebeam,DanielL.&AnthonyJ.Shinkfield,1985,SystematicEvaluation:ASelf-InstructionalGuidetoTheoryandPractice,Boston,MA,Kluwer–Nizhoff.
Tashakkori,Abbas&CharlesTeddue(eds),2003,HandbookofMixedMethods inSocial&BehavioralResearch,ThousandOaks,CA,Sage.
Taylor,Steven J.&RobertBogdan,1998, Introduction toQualitativeResearchMethods:AGuidebookandResource(3rdedn),NewYork,Wiley.
Thyer,BruceA.,1993,‘Single-systemsresearchdesign’, inR.M.Grinnell(eds),SocialWorkResearchandEvaluation(4thedn),Itasca,IL,F.E.Peacock,pp.94–117.
Trochim,WilliamM.K.& JamesDonnelly, 2007,The ResearchMethodsKnowledge Base (3rd edn),Mason,OH,ThomsonCustomPublishing.
Walter,Maggie (eds), 2006, Social ResearchMethods: An Australian Perspective, Melbourne, OxfordUniversityPress.
Webster’sThirdNewInternationalDictionary,1976,G.&C.Company,Springfield,MA.Yegidis,Bonnie&RobertWeinback,1991,ResearchMethodsforSocialWorkers,NewYork,Longman.Young,PaulineV.,1966,ScientificSocialSurveyResearch(4thedn),EnglewoodCliffs,NJ,PrenticeHall.
Index
100percentbarchart301–2
ABI/INFORM36accidentalsampling201actionresearch131activevariables71after-onlydesigns215Alkin,M.C.324alternatehypothesis85applicationsofresearch6appliedresearch9,10areachart305–6attitudinalscales167–76
calculatingattitudinalscore172difficultiesindeveloping169functions168inqualitativeresearch175inquantitativeresearch168relationtomeasurementscales175types170–5
attitudinalscore172–3attitudinalvalue174attitudinalweight174attributevariables71authenticity184
Babbie,E.178Bailey,K.D.10,74,83,245barchart298–300before-and-afterstudies107–10bias5,141,164,246bibliography403–7Black,J.A.63blindstudies126books,reviewing34Bulmer,M.8Burns,R.B.8,144,313
calculationofsamplesize209–12casestudies126categoricalvariable72causalchange
dependentvariable97–9independentvariable97–9total97–9
causevariable66–9Champion,D.J.63chancevariables96–9CINAHL36
classificatoryscale74–6closedquestions145,151–4clustersampling204–5codebook257–68codeofconduct241coding256–74codingqualitativedata277–88
codingquantitativedata255–77Cohen,M.R.63CohenandNagel63cohortstudies125columnpercentage295–7communityforums129,160,330communitysurveys330concepts
convertingintoindicators64–6differencebetweenconceptsandvariables63–4measurement73–7
conceptualframework40concurrentvalidity180conditioningeffect110confidentiality242,246confirmability185consent,244constantvariable73constructvalidity180consumers,opinions336contentanalysis277–88contentvalidity179continuousvariables73controlgroup117controlstudies117–19correlationalresearch9–11cost–benefitanalysis341–2credibility185cross-sectionalstudies107cross-tabulations294cumulativefrequencypolygon303–3
datacollectioninqualitativestudies159–63focus-groupinterviews160in-depthinterviews160narratives161oralhistories161
datacollectioninquantitativestudies138–59differenceinquantitativeandqualitativeresearch138methods138–59prerequisites159primarysources140–59secondarysources163–4usingattitudinalscales167–75
dataprocessing253–88coding256–88display291–308editing255–6inqualitativestudies277–88inquantitativestudies255–77
datasaturationpoint213databases36dependentvariables66–71dependability185DenzinandLincoln184descriptiveresearch10dichotomousvariables72disproportionatestratifiedsampling203–4DonnellyandTrochim185double-barrelledquestions155double-blindstudies126dropoutsfromanintervention334Duncan,O.T.74
editing,data255–6elevationeffect142equal-appearingintervalscale174ERIC36errorofcentraltendency142ethicsinresearch241–8
concept241–2participants244–6researcher246–7sponsoringorganisation247–8stakeholders243
evaluation,practice323–52consumer-oriented/client-centred342–3cost–benefit/cost-effectiveness341–2definitions324–5developingevaluationobjectives344–7ethics352focusofevaluation329goal-centred/objective-oriented342holistic/illuminative343impact/outcome337–41improvement-oriented329,343intervention–evaluationprocess327–8involvingstakeholders351–2philosophicalbase342–3processmonitoring/servicedelivery335–7program/interventionplanning330–2targetpopulationparticipation333–5types328–9undertakingevaluation343–51why325–6
evidence-basedpractice4experimentalgroup113experimentalstudydesigns113–26
after-only115before-and-after116comparative120control-group117cross-overcomparative123double-control119matched-control121placebo122replicatedcross-sectional124
expertpanel125expertsampling207explanatoryresearch9,11exploratoryresearch9,11externalconsistencyprocedures182–3extraneousvariables96–9
facevalidity179–80feasibilitystudy11feministresearch132FestingerandKatz18fishbowldraw200focusgroup127–8frameofanalysis274–6frequencydistributions274frequencypolygon302functionalanalysisstudies336
Gilchrist,L.244graphs297–307Grinnell,R.7,83
Guba,E.G.185Guttmanscale175
haloeffect143harm,causedbyresearch245HawthorneEffect141HEALTHROM36histogram298holisticresearch129HumanitiesIndex36Huxley,T.H.81hypothesis
alternate85ofassociation87characteristics84–5definition81–3ofdifference86errors87–8functions83null86ofpoint-prevalence87qualitativeresearch88–9types85–7
impactassessmentsafter-onlydesign115,338before-and-afterdesign116,339comparativestudydesign120–1,339doublecontrol119–20experimental-controldesign117–19,339interruptedtime-seriesdesign340matchedcontrol121–2placebo122reflexivecontroldesign339replicatedcross-sectionaldesign124,340
incentives,providing245independentvariables66–71indicators64–6,347–9information
inappropriateuse247misuse248sensitive156–8,245
informedconsent244inquirymode9internalconsistencyprocedures183–4internet37interpretiveparadigm14intervalscale74–7interveningvariables69–71intervention–development–evaluationprocess327–8interviewing144–5
advantages149–50constructingschedule156–7disadvantages150formsofquestions151–4formulatingeffectivequestions154–5schedule145,148structured145unstructured145versusquestionnaires148–9
journals,reviewing35–6judgementalsampling198,207
Katz,D.18Kerlinger,F.N.8,44,62,82,94,98,178
leadingquestions155Likertscale170–4limitations,research236–7LincolnandGuda185linediagram305–6Lipsey,M.W.324literaturereview31–42
reviewingprocedure33–40searchingexitingliterature34–5why31–3writing40–1
longitudinalstudies110Lundberg,G.A.7matching100maturationeffect109‘maxmincon’principle95–8measurementofconcepts64–6measurementscales73–7
interval75,77nominalorclassificatory75,76ordinalorranking75,76ratio75,77
MEDLINE36Meenaghan,T.M.44multipleresponses259multi-stageclustersampling205
Nagel,E.63narratives(informationgathering)161need-assessmentsurveys330negativestatements172neutralitems173nominalscales76non-discriminatoryitems171–2non-experimentalstudies113–14non-probabilitysampling197–8,206–8non-randomsampling197–8,206–8nullhypothesis86numericalscale171
objectives,formulating50,54observation140–4
non-participant141participant141problems141recording142–4situations141–2
100percentbarchart301–2open-endedquestions151–3operationaldefinitions55–7oralhistories161–2ordinalscales74–6outline(chapterisation)314–19
panelofexperts337panelstudies125paradigmsofresearch14–15participantobservation128,141participatoryresearch131–3percentages295–7piechart304–5
pilotstudy11placeboeffect122planningaresearchstudy23–6Poincaré18,73polytomousvariables72–3positivestatements172positivistparadigm14–15Powers,G.T.44prediction193predictivevalidity180pre-test158primarydata138primarysources138probabilitysampling198–205proportionatestratifiedsampling203–4prospectivestudies111pureresearch9–10purposivesampling207
qualitative–quantitativestudydesigns,differences103–5qualitativeresearch
analysis277–88comparedtoquantitativeresearch20sampling212–13,206–8
quantitativeresearchcomparedtoqualitativeresearch20frameofanalysis274
questionnaires145administering146–8advantages148coveringletter150–1disadvantages149versusinterviews148
questionsadvantages153–4closeended151–2disadvantages151–2formsof151–3formulating154–5open-ended151–2order158personal156–8pre-testing158–9sensitive156–8
quotasampling206–7
randomdesigns115–24numbers200–1randomisation114sampling198–206
ratioscale74–5,77rawdata255reactiveeffect109,395reconstructingthemainconcepts275referenceperiod106,111referencing320reflectivejournal130regressioneffect109–10reliability
concept181determining182–4factorsaffecting182inqualitativestudies184–6
reportwritingbibliography321outline314–19referencing320aboutavariable319–20
researchapplications4–6characteristics8–9definitions5,7–8objectives50,54operationalsteps18–23paradigms145problem44–7process18–27proposal217–37
reasonsfor1–4types9–14whatdoesitmean5,7–8writing313–14
researchdesigndefinition94functions94–5theoryofcausality95–9
researchjourney18–27researchproblem44
aspects45–7considerationsinselecting47–8stepsinformulating48–50,51–3formulationofobjectives50,54narrowing48–50operationaldefinitions55–7qualitativeresearch,in57–8sources45–7studypopulation55
researchproposal217–38contents218dataanalysis232–4ethicalissues231hypotheses226–7measurementprocedures230objectives224–6preamble220–2problemandlimitations236–7researchproblem224sampling231–2setting230structureofthereport235–6studydesign227–9workschedule237
researcher,ethicalissues246–7retrospective–prospectivestudies111–12retrospectivestudies111–12Rossi,P.H.324rowpercentage295–6Rutman,L.324
sampling192–214accidental207cluster204–5concept193design194disproportionatesampling203element194error195
frame194judgemental207non-probability198,206–8non-random198,206–8population193–4principles194–7probability199–206proportionate206purposive207inqualitativeandquantitativeresearch192quota206–7random199–206withorwithoutreplacement202–3size194snowball208statistics194stratified203–4systematic208–9terminology194unit194
saturationpoint213scattergram307schedule,work237Schinke,S.P.244searchengines37Selltiz,J.94secondarydata163–4secondarysources138–9semi-experimentalstudies,106,113–14sensitiveinformation156–8servicedelivery,evaluating335–7Shinkfield,A.J.324,342simplerandomsampling203snowballsampling208SocialSciencesCitationIndex35Solomon,L.C.324sponsoringorganisation,ethicalissues247–8stackedbarchart301stakeholders243statisticsandresearch308stem-and-leafdisplay304Stevens,S.S.74stratifiedrandomsampling203–4structuredinquiries11–12structuredinterviews144–5stub294studydesigns,philosophyoriented130–3
actionresearch131–2feministresearch132participatoryandcollaborativeresearch132–3
studydesignsinqualitativeresearch126–30casestudydesign126–7communitydiscussionforums129–30focusgroups/groupinterviews127–8holisticdesign129oralhistorydesign127participantobservation128–9reflectivejournallog130
studydesignsinquantitativeresearch105–24before-and-after107comparativestudydesigns120–1controlstudies117–19cross-overcomparativedesigns123cross-sectional107
double-controldesigns119–20experimental113–14longitudinal110matched-controldesigns121non-experimental113–14placebodesign122prospective111–12random114replicatedcross-sectionaldesigns124retrospective111–12retrospective–prospective111–12semi-experimentalorquasi-experimental113–14
studypopulation55–6,193definition55ethicalissuesaffecting244–5evaluatingparticipation333–5
Stufflebeam,D.L.324subjectarea46–7subjectivity5summatedratingscale170–4systematicsampling208–9
tables292–5targetpopulation193text292thematicwriting41theoreticalframework38–40Thurstonescale174Thyer,B.A.94time-frame237time-and-motionstudies336totalpercentage297transferability185treatmentgroup113trendcurve305trendstudies124–5Trochim185Twoomey,B.G.44TypeIerror88TypeIIerror88
unitofmeasurement62unstructuredinquiries11–12unstructuredinterviews144–5
validity178–81concept178–9concurrent180construct180–1credibility184–5dependability184–5face179–80predictive180inquantitativeresearch179–81inqualitativeresearch184–6transferability185types179–81,184–6
variables62–73active71–2attribute71–2categorical72–3chance97–8change66,68andconcepts63–4
connecting69–70continuous72–3definition62–3dependent66–70dichotomous72–3extraneous66–8independent66–70intervening69–70measurementscales73–7outcome68–71polytomous72–3types66–73
workschedule237workingdefinitions55–7
Young,P.V.63