developing a public online learning environment for crisis ...gala.gre.ac.uk/19713/2/19713...

DOI: 10.4018/IJISCRAM.2017040102

International Journal of Information Systems for Crisis Response and ManagementVolume 9 • Issue 2 • April-June 2017

Copyright©2017,IGIGlobal.CopyingordistributinginprintorelectronicformswithoutwrittenpermissionofIGIGlobalisprohibited.

Developing a Public Online Learning Environment for Crisis Awareness, Preparation, and ResponseLiz Bacon, Faculty of Architecture, Computing and Humanities, University of Greenwich, London, UK

Lachlan M MacKinnon, Department of Computing and Information Systems, University of Greenwich, Old Royal Naval College, London, UK

Avgoustinos Flippoupolitis, Department of Computing and Information Systems, University of Greenwich, London, UK

David Kananda, University of Greenwich, London, UK

ABSTRACT

Thisarticledescribesthedesignanddevelopmentofanonlineimmersivelearningenvironmentfocusedonenhancingthegeneralpublic’sawarenessof,andpreparationfor,crisissituations.Thisresearchhassoughttoanswerthequestion“Isitpossibletodevelopatimelinebasedimmersiveandengagingtrainingenvironmentformassself-studyeducationincrisispreparedness?”ThesystemdevelopedisbasedonthePandora+trainingenvironmentandintegratesoriginalcollaborativeEuropeanresearchworkcarriedoutoneLearningandCrisisManagementoverthelasttenyears.TheresearchreportedherenotonlydescribesthedesignofthePandora+trainingenvironmentbutalsotheoutputsfromapilottrialinLisbonrunbythePOP-ALERTEUFP7project.Whereappropriate,theresultswerealsocomparedtothosefromalargeEUsurveyoncrisispreparednessandattitudes,alsoundertakenwithinPOP-ALERT.Theresultsofthisarticlehaveresultedinanoriginalandinnovativesystemthathassignificantpotentialtotransformtheeducationofthepublicindisasterpreparedness.

KEywORDSCrisis Preparedness, MOOC, Online Learning, Public Self-Study

INTRODUCTION

For the past ten years the lead authors have been involved in a research agenda encompassingthe provision of, and support for, training for all stakeholders involved in crisis management.Predominantly,ourwishistomakewell-formedandeasilyaccessibletrainingmaterialsavailableatrelativelylowcost,toimprovethequalityofpublicandprofessionalresponseincrisissituations.Toenablethiswork,theauthorshavebeeninvolvedinanumberofpubliclyfundedresearchprojectsrelatedtothisresearchagenda,andthispaperbringstogetheroutputsfromthreeofthoseprojectsinthedesignanddevelopmentofanonlinevehicleandtrainingmaterialstohelpdeveloppopulationawareness.Chronologically,thethreeprojectsare,thePandoraproject,thedCCDFLITEproject,bothofwhicharedescribedlaterinthepaper,andfinallythePOP-ALERTproject,whichisthemainfocusofthisworkandisabouttocomplete.POP-ALERTisanEUFP7projectinvolvingelevenpartnersfromsevencountriesacrossEurope.Itsfocusisthepreparationofsocietiesandpopulationstocopewithcrisesanddisastersinarapid,effectiveandefficientway.ThePOP-ALERTteamhasundertaken

18


19

athoroughreviewoftheliteratureonapproaches,populationbehaviours(includingwillingnesstoprepare),firstreactionstrategies,awarenessofrisketc.Thefocusofthisworkisnotonlyonlocalpopulationsbutalsovulnerablegroups,suchastourists,expatriates,theelderlyandrefugees,andtheeffectivenessoftheuseofmessages,audiblealarms,pictogramsetc.onthesepopulationgroups.ThePOP-ALERTprojecthasgeneratedaframeworkencompassingavarietyoftoolsandtechniquestoenhancepopulationawareness, realisedasanonlinedashboard,andconcludedwith localanddistributedfieldstudiestotesttheeffectivenessofthisframework.

TosupportthefieldstudiesinPOP-ALERT,theauthorsutilisedabespokeversionofPandora+(1Baconetal.,2015),whichistheenhanceddevelopmentofaproductcalledPandora(Baconetal.,2012;MacKinnonetal.,2013),oneofthekeyoutputsfromanEUFP7project,whichranbetweenJan2010andMarch2012.Pandora+isanimmersive,richmultimedia,trainingenvironmentinitiallydesignedtoproviderealistictrainingforstrategiclevelcrisismanagers,who,intheeventofacrisis,needtoworktogethertocomeupwithaplanofactionandtakedecisionsasthecrisissituationunfolds.ThePandora+systemisnothoweverspecific tocrisismanagement, it isasophisticatedenvironmentintowhichawidevarietyofscenariosfromanydomaincanbeuploadedandexecuted.Ithasaneventnetworkapproach,whichpresentstraineeswithaseriesofeventsthatoccurwithinaspecifiedtimelineandrequiresthemtomakedecisionsaboutwhattodoatspecificpointsonthattimeline.Thetimelinethenhasthecapacitytobranchthescenariodependingontheanswerprovidedbythetrainee(s).Anexampleofadifferentdomaincouldbeatrainee’sanalysisoftheriseandfallofthestockmarketinrelationtospecificevents,andmakingdecisionsaboutstocksandsharestopurchase/selletc.

ThePandora+environmentwasutilisedinafieldstudyinPOP-ALERTthatwasusedtotrainmembersofthegeneralpublicinLisboninFebruary2016.Thescenariofocusedonanearthquakeasanexampleofanaturaldisaster.OnepurposeofthetrialwastotesttheuseofPandora+andthePOP-ALERTdashboardonasmallsampleofthepopulationasapilot,beforerunningitasaself-studymassiveopenonlinecourse(MOOC)(2Baconetal.,2015),withmembersofthegeneralpublicabletoregisterandfollowthetrainingcourseatatimesuitabletothemandforadurationoftheirchoosing.

Inmakingtheshiftfromtrainingprofessionalcrisismanagers,toaMOOCusedbythepublic,changestothePandora+environmentwererequired,andthisincludedtakingonboardthebehavioursandapproachestotrainingbythegeneralpopulation,whenengagedwithself-study.ThesechangescamefromlessonslearnedfromtherecentdeliveryofaMOOConEntrepreneurshipandInnovation,inwhichsomeoftheauthorswereinvolved,aspartofanErasmus,Life-longlearningprogrammeproject,calleddCCDFLITE-distributedConcurrentDesignFrameworkforeLearninginITEntrepreneurship(FLITEforshort),whichranfrom1stOct2010to30thSeptember2015(2Baconetal.,2015).

Combiningtheoutputsofthesethreeprojectsprovidesauniqueandinnovativesystemthataimstoanswertheresearchquestion“isitpossibletodevelopatimelinebasedimmersiveandengagingtrainingenvironmentformassself-studyeducationincrisispreparedness?”.

Therestofthispaperisstructuredasfollows:itfirstofalldiscussessomeofthekeyissuesinattitudesandtrainingofthegeneralpopulationincrisisresponseandtheneedtotakeintoaccountbehaviours,attitudes,willingnesstoprepareetc.ItthendiscussesthefactorstotakeintoaccountfromtheeducationalperspectiveandtheresearcharoundonlineeducationandMOOCs.Finally,itfocusesonthedesignanddevelopmentofadistributedversionofPandora+requiredtotrainthegeneralpopulationincrisispreparedness,reportsontheresultsfromtheLisbonpilot,andthendrawssomeconclusionsandoutlinesfuturework.


20

POPULATION PREPAREDNESS: ATTITUDES AND BEHAVIOURS

Whenacrisisoccurs,andapopulationisalerted,peoplegothroughseveralstagesintheircognitivethoughtprocesseswhicharereasonablyconsistentacrossthepopulationregardlessofwhetherthereis an immediate threat or not. The sequence of thought is typically as follows: receive, believe,personalise,respondandconfirm(Paton2006;Mileti&Peek2000;Perryetal.1982)andthesecanoccurquickly foran immediate threatorovera longerperiodof time ifacrisis situation isunfolding.Intermsofpersonalisation,peopleaskthemselveswhatitmeansforthem,andthatisthekeyquestionpeoplewillseekanswerstoduringacrisisinordertoassesstheirpersonalriskgoingforward.Researchhasshownthatthegreatertherelevancetooneself,thegreatertheperceptionofrisk(Dow&Cutter,1998;2000).

Therearehowevermanyfactorswhichinfluenceaperson’sdecisionwhetherornottoprepareforapossiblecrisis,andtheirbehaviourduringacrisis.Keyfindingsaresummarisedbelow:

• Themannerinwhichinformationonadisasterorcrisisispresentediskeytopeople’sreaction,forexamplehowcertaininformationis,severity,immediacy,proximity,natureoftheevent,orcertaintyofapotentialevent(Seydlitzetal.,1994).Itseemsthatpeopletendtooverestimateriskforrareeventsandunderestimateriskforseeminglyfrequentevents.People’sbeliefsaboutahazardcanbeamixoffactualknowledge,misinformation,myths,hypothesis,beliefsabouthumanbeings,trustinauthoritiesetc.

• Womenaremorelikelytopreparebecausetheyhaveahigherperceptionofrisk(Fothergill,1996citedinPaton,2006),however,thisisaffectedbyculturalissues,decision-makingculturesofthefamily,etc.MenaremorelikelytotuneintoTVandradio,whereaswomenaremorelikelytousesocialnetworksandcallpeople(Bagrowetal.,2011).

• MostpeoplehaveasignificantexpectationthatlocalauthoritiesandcentralGovernmentwillpreparefor,detectandmanagedisasters,andtheirrecovery.Theyexpectclearcommunication,andfortheauthoritiestolookafterthemintheeventofadisaster.

• Confidentpeople,orpeoplewithhighself-efficacy,tendtobelievethatathreatwillnotaffectthem,andtheytendtothinkpositivelyaboutathreat.Theyarealsolesslikelytopreparefordisasters.Theirbeliefintheirowncopingstrategiesmaymitigatetheirsenseofrisktodailyroutineandproperty.

• Whethersomeoneisresidentinanareawheretheyperceivethatadisastermightoccur.• Peoplewhohavemoreexperiencesofdisastersareinclinedtohaveanoptimisticviewabout

theircapacitytosurvivebutarealsomorelikelytoadapttheirbehaviour.• Thecostandinconvenienceofpreparationisafactorinpeoples’decisiontoprepare.

Therearehowevermanyvulnerablegroupsatatimeofdisaster,andithasbeenshownthatthelesscontroladisadvantagedpersonhas,thelesslikelytheyaretoprepare(Legates&Bidel,1999).Vulnerable/disadvantagedgroupscanbesummarisedasfollows:

• Transientpopulationssuchastourists,travellers,migrantworkers,clandestinecommunitiesandoverseasstudents.Ingeneral,theyareconsideredatadisadvantageduringdisasters(Quarantelli,1994)astheyarenotintegratedintolocalcommunities.Thisdisadvantagemaybewhytransientpeopleareamongsttheearliesttoevacuatewhendisasterbecomesaprospectbutarelesslikelytobepreparedortohavetakenprecautionaryaction.Theyarealsotheleastlikelytoseekhelpfromauthorities(Lindell&Perry,2004).

• AlistofvulnerablegroupsfromtheVulnerablePopulationsOutreachModelbasedonEnglish-speakingnations(Klaimanetal.,2010)wasdevelopedtoidentifypeopleatriskduringdisasters.These include: non-English speakers, visually impaired, hearing impaired, isolation due toeconomiccircumstances,isolationduetomedicalcircumstances,lowlevelsofliteracy,homeless,


21

prisoners,belowthepovertylevel,peoplewhodependonserviceanimals,children,olderandfrailadultsandunemployedpeople.

• Minoritygroups(gender,religion,ethnic,disability,sexuality,income)areoftenmorevulnerableforavarietyofreasonse.g.differencesintermsofresources,inappropriatenessofgenerictrainingprogrammes,andtheyhavebeenfoundtobelesslikelytoevacuate(Perryetal.,1982),astheyexhibithigherlevelsofscepticismaboutwarnings.Otherreasonswhyminoritycommunitiesaremorevulnerableincludesocial,economic,cultureandlanguagebarriers,lowerperceivedpersonalrisk,distrustofmessages,lackofpreparationandprotectiveaction,andrelianceoninformalsourcesofinformation.

• Disasterspectatorsandvolunteerswhotraveltoadisastersiteforcuriosity,adventure,tohelp,towitnessacontinuingdisaster,etc.Theseincludenewsmedia,peoplewantingtowitnessanon-goingcrisise.g.stormchasers,peoplewantingtohelpbutnotknowinghow,etc.Peoplecanputthemselvesinvulnerablesituationsintentionallyandunintentionally,andauthoritiesneedmechanismstodealwiththefullrangeofthesepeople.

AspartofthePOP-ALERTprojectasurveywasconductedacrossEuropeofcitizen’sexperiencesofdisasters,theirperceptionofthreatandrisk,theirexpectationsofauthoritiesandtheirattitudestopreparation.Thesurveywasansweredby1612participantsandwasissuedin6languages,thebreakdownbeing:610English-languageparticipants,483French-languageparticipants,62Dutch-languageparticipants,88Greek-languageparticipants,145German-languageparticipants,and224Portuguese-languageparticipants.Thefullsurveyresultsarereportedelsewhere(Filippoupolitisetal.,2015),howevergiventhefocushereisontrainingthepopulationinpreparedness,twokeyquestionswereaskedaboutpeople’swillingnesstoprepareandtheresultswereasfollowsinTables1and2:

Forthosewhodidnotintendtoprepareorintendedtoprepare(firsttwoanswers)theywereaskedafollowupquestion.Notethattheycouldtickmultipleanswersandtheaveragerespondentticked1.66answers:

Table 1. Initial Survey Question Results

Whichstatementbestrepresentsyourpreparednessforadisaster?

Idonotintendtoprepare 15.80%

Iintendtoprepare 42.96%

Ijuststartedpreparing 19.88%

Iamprepared 21.36%

Table 2. Follow-up Survey Question Results

Whyhaveyounotprepared?

Idon’tknowwhatIshoulddo 35.94%

Ididn’thavetime 16.80%

Itcoststoomuch 6.82%

Idon’tthinkitisimportant 9.03%

Idon’tthinkitispossible 11.37%

Emergencyserviceswillhelpme 13.90%

Other 6.13%


22

So,thegoodnewsisthatifweassumethatthe15.85%ofthisgroupthatsaythattheydon’tthinkitisimportantoritcoststoomuchareunwillingtoprepare,wecanalsoreasonablyassumethattheotherresponsesareopentonegotiation.Thismayprovideameansofaddressingatleastsomeofthe15.8%ofthepopulationwhodonotintendtoprepare.However,pragmaticallywecantargetthe42.96%ofthepopulationthatmaybewillingtopreparebutareclearlyunsurehoworareconcernedovertimeforexample,andalsothe19.88%thathaveonlyjuststartedpreparing,whichprovidesasignificantproportionofthepopulationthatarepotentiallyopentotheprovisionoftraining.

Thefactthatonly21.36%oftherespondentsareactuallypreparedforacrisissituationgivesaclearindicationofthescaleoftheproblem,especiallywhenyouconsiderthattherespondentstothesurveywereinevitablythosewhohadagreaterlevelofinterestin,orawarenessof,theissues.Whilstwemightseektomakeinroadsintothegroupwhodonotintendtopreparethroughpublicinformationprogrammesortargetedapproaches,therealityisthattheseareonlylikelytobecomeeffectivewhenthemajorityofthepopulationisprepared,sofocusingonthosewhoarewillingtoprepareislikelytobemorebeneficialinthelongterm.

It is clear from the survey that informationonhow toprepare isnotperceived tobe easilyavailabletothepopulationorisnotseenasapriority.However,regardlessofplanningandpreparation,respondentscitedmanyreasonswhytheymaydelayevacuation,includingseekingfurtherclarificationfromofficialsources,caringforrelatives,evacuatingpets,andgatheringpersonalbelongings.So,thedesignofanypreparationtrainingneedstotakeintoaccountissuesthatmightcausesuchdelaysandprovideinformationandsupportinresolvingtheseissues.However,itwasnotedintheGermansurveythatthehighestpercentageofparticipantsthatwoulddelayevacuationareamongthosewhodonotintendtoprepare,whichclearlycompoundstheproblem,andprovidesinformationforemergencyservicesratherthanthedesignofpreparationtraining.

Wecanalsodrawonahigherlevelofwillingnesstopreparewhentravellingabroad,oronpublictransport,expressedinthesurvey,asevidencethatproperlytargetedanddesignedawarenessandpreparationtraininghasthepotentialtobeeffective.Infact,thesurveysuggeststhatmakingsuchtrainingreadilyandeasilyavailable,sanctionedandsupportedbylocalauthorityandemergencyservices,couldhaveasignificantimpactonpublicawarenessandlevelsofpreparation.

ONLINE EDUCATION AND MOOCS

InEurope,theliteratureonthebenefitsoftrainingthepublicandthebenefitsoftrainingfirstrespondersissparseandsuggeststhatlearningmaynotresultinmoreprotectiveaction(Leonardetal.,2008).Forexample,halfofthepeoplewhohadattendedtraininginonestudysaidtheyknewhowtheywouldbeinformediftherewasadisaster,howeverathirddidnotrecognisethealarmsignalafterthetraininghadbeengiven.Lalo(2000)observedthatthereisnotnecessarilyadirectlinkbetweenpeople’sactualbehaviourandwhattheylearned,andofcourselearningfadesovertime.Heconcludesthateffortsdesignedtopromote“socialunity”aresignificantinshapingresponsesbeforeanydisaster.Thispredicatesregularandeasyaccesstotrainingforreinforcementoflearning.

Itishoweverclearfromourresearchthatthevastmajorityofthepopulationarewillingtopreparebutasignificantproportiondonotknowhow.Therearemanywebsitesprovidingadviceondisasterpreparationandassociatedkitsforbothhumansandpets,howeverfromtherespondents’answerstooursurveyquestions,itwouldappearmostarenotawareofthem,astheyareunclearhowtoprepareandthesewouldhavebeenausefulresource.Interestingly,therehasbeenatleastoneattempttoofferdisasterpreparationintheformofaMOOC,whichhasrunontheCourseraMOOCplatformseveraltimes,howeverMOOCsingeneraltendtohaveverypoorsuccessratessoitcouldbearguedthattheymaynotbethebestapproach.However,theydoremainattractive,astheyhavetheabilitytoengagealargeaudienceandtraditionalapproachesmaynotbesufficientgiventhescaleofthetrainingrequiredforthepopulationsidentified,ortheabilityofthosepopulationstolocatetraining.


23

Fromourexperience,MOOCsdohavetheirchallenges,thepedagogyisimportantandmanytutorsapproachonlineeducationinthesamewayastraditionalteaching,withtheendresultbeingapoorerversionofaclassroom-basedapproachputonline.InaMOOC,theroleofateacherhastochangetobecomemoreofafacilitatorthananinstructoras,giventhenumberofstudentsinvolved,personalisedtuitionbyaninstructorisimpossible.However,personalisationofthelearningenvironmentisofcoursepossible,andmoresothaninaclassroom-basedapproach.InrecenttimestherehasbeenamovemoretowardscMOOCs,whicharebasedonaconstructivist,student-centricapproachtolearning,whichsupportsstudentindependentlearningfacilitatedbyaninstructor.Irrespectiveoftheapproach,asnotedabove,MOOCssufferfromhighdropoutratesandthereareanumberofreasonsforthis,suchasastudent’sabilitytocopewithlesssupportfromtheinstructorinalessstructuredenvironmentwhichrequiresthemtobecomeamoreindependentlearner,managetheirownworkloadsandtimeetc.Researchsuggeststhatifastudentistobesuccessfulthenitiscrucialthattheyfeelpart of a learning community (Ellis 2001; Bernard 2000). However, many MOOC learners canstrugglewithaMOOConlinecommunityfeelingoverwhelmedbythevolumeofmaterial,choiceoflearningmaterialsavailable,thecomplexityoftheenvironments,thetoolsused,thenumberofparallelconversationstakingplace,networksthatformanddisbandastimepasses,etc.(Kopetal.,2011).Whilst thefacilitiesavailableondifferentMOOCplatformsdiffer, theyallhavea typicalcommoncore,forexample,aplacewherelearnermaterialsarelocated,linkstoreadingmaterials,videostowatch,discussionforafordebates,quizzesetc.However,bycomparison,playersengagedinimmersiveonlinegamingenvironments,whicharealsoopentomassivenumbers,MMORPGs(MassiveMulti-PlayerOnlineRole-PlayingGames),achievefarhigherlevelsofengagementandreturn.Ingeneral,whilsttheremaybeshort-livedimmersivecomponentswithinaMOOC,theyaregenerallynotfullyimmersive,engagingexperiences.So,ourapproachhereistocombinethebestofbothworlds,i.e.toprovideanengagingandfullyimmersive,richmultimedia,trainingenvironmentasaMOOCplatform.

Whilsttheexperiencewillbedesignedtobeengaging,ourresearchalsoraisedtheissueofhowwemeasuresuccesswithinaMOOC.Thetraditionalmeasureistogaugesuccessbytheteacher’sexpectations,whicharetraditionallythatallstudentswillcompletethecoursetotheendandsubmitsomefinalassessmentasproofofcompletion.However,whenweaskedthestudentsinourMOOCtheirintentionsatthestartofthecourse,closeto50%madeitcleartheyhadnointentionofcompletingthecourse(andthoseweretheonessufficientlyengagedtoanswerthequestionnaire).Theymadeadecisionatthestartwhattheyexpectedtogetfromit,e.g.manywerelookingforsomespecificpieceoflearningandjustplannedtodipingetthesnippetoflearningtheywanted.So,measuringsuccessbycompletionisnotappropriateinMOOCsasonecouldclearlyarguethatifthestudentgotfromitwhattheywanted,thenthatshouldbecountedasasuccess.

So, in order to provide mass online disaster preparation education for the general public, adecisionwastakentodevelopthePandora+environmentasadistributed,online,self-study,immersiveenvironmentthatcouldberunasaMOOCplatform.ThefollowingsectionprovidesdetailsofthedesignanddevelopmentofthedistributedonlineversionofPandora+.

PANDORA+ DESIGN AND DEVELOPMENT

ThissectionfocusesonthechangesthathavebeenrequiredtothePandora+trainingenvironmenttobeabletodelivertrainingtothegeneralpublicintheformofaMOOC.

AkeyfocusofthedesignofthePandora+environmentisthemanagementoftraineestress,asresearchhasshownaclearlinkbetweenstressandcognition(Salesetal.,1996).Stressreducesaperson’spsychologicalcapacity.Itcanreducetheirabilitytoundertakeprotectivebehavioursincludinginformationseekingandprocessingthatinformation(Vihalemmetal.,2012).Inshort,aperson’sdecision-makingabilitiesareaffectedbytheirlevelofstress,withthepotentialforsomeonetodevoteinsufficient timetoconsideringalternativesolutionsandalso toconsider theminadisorganised


24

manner,andtomakedecisionsbeforeconsideringallthepotentialinformation,etc.Forthisreason,thePandora+environmentwasdesignedwithanumberofcontrolstosupportthetrainerinmanagingthestressofeachtraineeinordertocreateasrealisticacrisisenvironmentaspossible.

Thetypeofcontrolsthetrainerhasinclude:

• Theabilitytodynamicallychangeascenarioduringatrainingsessiondependingonhoweachtraineeisresponding

• Theabilitytospeeduporslowdowntheexecutionofascenario• Theabilitytoaddneweventstothescenariotocreatemorepressureonthetrainees,e.g.inform

themthatalorryhasjustcrashedinalocaltunnelanditscargohasnowpollutedthewatersupplyofthelocaltown

• Theabilitytoroleplayacharacterinthescenariotoallowthetrainertoexplorethethinkingoftraineegroupandthusallowawiderrangeofoutcomesfromthetrainingevent.

ThePandora+trainingenvironmenthassomeautomatedfeaturestohelpmanagethestressofthetrainees,includinganinitialquestionnairetoassessbehaviouraltraitssuchasself-efficacy,personalitytraits,stressandanxiety,leadershipstyleetc.,andothermeasuressuchasbiometricinputsandself-reporting.Asaresult,thesystemcanautomaticallyaddeventsfromitsrule-basedenginetotryandincreaseordecreasethelevelofstressofoneormoretrainees.However,thisfeatureofthesystemwasnotbeingusedinthepilottrialstobeginwithduetothepracticalityofcapturingbiometricdata,andalsothenatureofscenariodevelopmentforindividualswithdifferentexperiencelevels.Priortothepilot,thesystemhadonlybeentestedoncrisismanagementprofessionals,notthegeneralpublic,soitwasfeltthatmoretrialswererequired,postpilot,beforeattemptingtogeneratethestressofarealisticcrisisscenariointhegeneralpublic,howeverthisisanimportantareaofresearchthatwillberequiredinfuturetodelivermorerealisticcrisispreparationtraining.

BoththeoriginalPandoraandcurrentPandora+systemswereinitiallydesignedtobeusedonlyinanenvironmentwithatrainerpresent.Theywereboththereforedevelopedtorunoveralocalareanetworkwiththeclientinstalledonthetraineecomputersandtheserverinstalledonthetrainercomputer (seeFigure1).However, this isnotpractical forusewith thegeneralpublic and soanumberofchangestothePandora+trainingenvironment,toenableittobeusedasaMOOCplatform,wererequired.Inordertoretaintherichmultimedia,immersivenatureoftheenvironment,itwasstillrequiredtobefocusedaroundascenarioincorporatingawiderangeofmultimediaelements,withasequenceoftime-driveneventsdemandingregulardecisions/inputfromthetrainees,butitcouldnolongerbedrivenbythepresenceofahumantrainer.Thesystemhadtobeofferedasawebservicesothatthetraineeenvironmentcouldberuninastandardwebbrowser(seeFigure2),ratherthanrequiringanyspecialistsoftwaretobeinstalledonthetraineemachine.Additionally,toreducerequirementsontheuserstolearnanunfamiliarinterfaceinordertousethesystem,theinterfacewasdesignedtobesimilartotheinterfacedesignofMicrosoft®Office,beingpossiblythemostfamiliarinterfaceworldwide.Sinceitwasnotpossibletopredictwhatdevicesuserswouldwishtouseforthetraining,aresponsivedesignapproachwasadoptedtoenabletheinterfacetobeviewedonmobiledevicesifnecessarybutnotingthatthevolumeofinformationtoviewonamobileatanyonetimewouldbeparticularlychallenging.Atthepilot,laptopmachineswereusedforallparticipants.

Anothermajorchangetothedesignwastoparameterisethelanguageusedforalltheinterfaceandhelp information.Thiswasnecessary toenhanceengagementwith thegeneralpublicacrossEurope.Obviously,theactualscenarioinformationuploaded,includingthemultimediaassets,e.g.anewsvideoofadisasterreport,wassubtitledforthelocallanguageofdelivery.

Takingonboard the lessons learnedfromtheMOOC, the lengthof the trainingscenario issignificant.Manymorepeoplewill find time for fiveminutesof training than thirtyminutesoftraining,howevermanymaywishtodomorethanfiveminutesandsobite-sizedchunksoftrainingareimportanttooffertothepublic,dividedupbytopicsofkeyinteresttothemsotheycanchoose


25

theorderinwhichtheyengage.However,forthepilot,withacaptiveaudience,a35-minutescenariowasutilised.

OneofthekeyfeaturesofthePandora+systemofnoteistherecordingfeature,whichlogsallcommunicationanddecisionsmadebyagroupoftraineesforuseindebriefingsessionswiththetrainerafteratrainingevent,eitherasagrouporanindividual.Whilstthereisnotrainerinthisdistributedpilotversion,becausethepublicwereaskedtoworkthroughthescenariointhesameroombutattheirownpacetherecordingtoolisstillavaluablefeature,inthatitgathereddataondecisionstakenbythepublic.Whenusedforthelargetrialswiththepublic,itwillproviderichinformationabouthowpeoplemayperformintheeventofacrisis.Thisdatawillbegatheredandanalysedonanon-goingbasistoenhancethetrainingmaterials.Combinedwiththedemographicdatacapturedwhenusersregistertousethesystem,thiswillalsoprovidearichsourceofdatathatcanbeanalysednotonlytoimprovethequalityofthetrainingmaterials,butalsotodetermineuserappetitefortraining,preferencesfortypeandnatureofinformationandsupportforpreparation,andlevelsofknowledgeandpreparation.Potentially,thiscouldbearichsourceofdataforfuturecrisispreparationinitiativesandmayofferavehiclefortrialsofsuchinitiatives.

RESULTS FROM LISBON TRIAL

Inthissectionwepresenttheresultsofareal-worldtrialweconductedinLisbon,PortugalinFeb2016.ThegoalofthetrialwastoassesswhetherthePOP-ALERTdashboardcouldbebeneficialto

Figure 1. Trainer Interface: Showing Scenario Text Highlighted and Associated Timelines


26

thewaypopulationsrespondtodisasters.ThePOP-ALERTdashboardisoneofthekeyoutputsoftheprojectandisdesignedtoprovidetoolsforthegeneralpopulationtoprepareforvariousdisastersandcrisis,receivealerts,understandtheir localalertingsystem,receivenotificationsfromsocialmedia,etc.Aseconddimensiontothedashboardistoprovidetoolsforpublicauthoritiestohelpthemworkbetterwiththegeneralpopulation,includingtipsonhowtocreateaninformativewebsite,howtousesocialmediaforalertingandpreparedness,howtoplananawarenessraisingcampaign,howtotrainthegeneralpopulation,etc.However,thefocusofthepilotwastotestitsusewiththegeneralpopulation.

Toachievethat,weusedthePandora+systemtomodelanearthquakescenariotakingplaceinLisbon.Thetotalnumberofparticipantswas65andtheyweredividedintwogroups:acontrolgroupcomposedof29participantsandanexperimentalgroupinvolving36participants.BothgroupsusedthePandora+trainingenvironmentandwereaskedtoprovideresponsestoquestions,forexampleaboutactions theywould takeas thedisasterscenariounfolded.Thedifferencebetween the twogroupswastheirorderofexposuretothedashboard.Morespecifically,thecontrolgroupwasaskedtofirstusePandora+andgothroughtheemergencyscenariowithouthavingused,orknownabout,theexistenceofthePOP-ALERTdashboard.Ontheotherhand,theexperimentalgroupfirstwentthroughatrainingsessionthataimedtointroducethemtothefeaturesandthefunctionalitiesofthePOP-ALERTdashboard(Figure3).Afterthis,theparticipantsusedthePandora+systemtoworkthroughtheearthquakescenarioandanswerthequestionsillustratedinthefollowingtables.Beforeproceedingwiththeanalysisofthetrialdata,itisimportanttonotethattherecordingmechanismdescribedabovecapturedtheresponsesoftheuserswhilealsoprovidingthemwithanimmersiveexperienceinvolvinganevolvingdisastersituationrelatedtoanearthquake.Thisexperienceincludedtheuseofstreamingvideotoannouncetheemergencytothepublic,displayofsafetypictogramsthattheuserhadtoidentifyandawiderangeofquestionsonemergencyawareness,preparednessandactionstheywouldtakeinresponsetotheunfoldingevents.

Figure 2. Trainee Interface: Modeled on MS® Office Look and Feel


27

Someoftheanswerstothequestionsfrombothgroupsarediscussedbelow.AswecanobservefromTable3, theexperimentalgrouphasgivenmoredetailedanswerswithrespect totheirfirstreaction.Incontrastto13.79%ofthecontrolgroupwhostatedthatseeingthenewswouldmakethemconcerned,only2.78%oftheexperimentalgroupmemberssharedthisview.Althoughtheirmostpopularanswerwasthattheywouldkeepsafe,whichislackingdetail,thenextmostpopularanswerwasthattheywouldprepareanER(emergencyresponse)kit.Thelatter,alongwithanswerssuchaslisteningtotheradioandbuyingfood,illustratethatusingthedashboardhashelpedtheparticipantshavingamoreappropriatefirstreactiontothenewsofanevolvingdisaster.

Table 3. Responses to question on “What was your first reaction to seeing the news regarding the earthquake?”

First reaction to seeing the news

Control Experimental

Keepsafe 10.34% 13.89%

Concerned 13.79% 2.78%

Checkonpeoplenearby 10.34% 2.78%

PrepareERkit 0.00% 8.33%

Shocked 3.45% 5.56%

Alarmed 3.45% 2.78%

ListentoTV/radio 0.00% 5.56%

Buyfood 0.00% 2.78%

Keepcalm 3.45% 0.00%

Offerhelp 0.00% 2.78%

Figure 3. The POP-ALERT Dashboard Home Page


28

ThebeneficialeffectofhavingbeenintroducedtothedashboardisdepictedinTable4.Aswecanconfirm,theexperimentalgroup’smostpopularanswerwasthattheywouldleavetheirpetstoasafeplace,anactionthatcoincidesthesuggestedonefoundinthedashboard.Onthecontrary,thevastmajorityofthecontrolgroupstatedthattheywouldkeeptheirpetswiththem,whichisnottherecommendedaction.Table5alsoillustratesthebenefitsofthedashboard,capturedviatheinteractiveinterfaceofPandora+.Morespecifically,themajorityofthedashboardusersagainstatedtheywouldleavetheirpetsinasafeplaceandonly5.56%thattheywouldtakethemwiththem.

ThereislittledifferencebetweentheresultsinTable6.Overall,bothgroupshavegivensimilaranswers regarding their reactionwhen indoors.Thesamestands for thecaseofoutdoors.Theseresultsindicatethatdeprivingtheuseroftheresourcesofthedashboarddidnothaveaneffecton

Table 4. Responses to question on “If you have a pet: What would you do with them during the earthquake?”

If you have a pet: What would you do with them during the earthquake?


Keepwithme 51.72% 19.44%

Leaveinsafeplace 10.34% 33.33%

Putonaleash 0.00% 19.44%

Calmthemdown 3.45% 5.56%

Nothing 0.00% 5.56%

Setfree 6.90% 0.00%

Abandon 0.00% 2.78%

Table 5. Responses to question on “If you have a pet: What would you do with them after the earthquake?”

If you have a pet: What would you do with them after the earthquake?


Checkonthem 10.34% 5.56%

Takewithme 10.34% 5.56%

Calmthemdown 3.45% 5.56%

Putonaleash 0.00% 2.78%

Leaveinsafeplace 10.34% 33.33%

Table 6. Responses to question on “If you are directly affected by the earthquake: What actions would you take when you are indoors?”

If you are directly affected by the earthquake: What actions would you take when you are indoors?


Gounderfurniture/doorframe 51.72% 52.78%

Gotosafeplace 13.79% 16.67%

Crouch 6.90% 2.78%

Switchoffgas/electricity 3.45% 0.00%


29

theiranswers.Inotherwords,participantshadhighlevelsofawarenesswithrespecttothesetwotypesofemergencysituations.

Theresults,however,forthecasewheretheemergencyfindstheparticipantsinsideavehiclearesignificantlydifferent.Themajorityofthecontrolgroup(27.59%)statedthattheywouldleavethecar,anactionthatisnotrecommendedintheeventofanearthquake.Theanswersfromtheexperimentalgroupalmostexcludetheoptionofleavingthecar(2.78%)andthemajoritypreferredtostayinthecar(38.39%)whichindicatesthattheyhavebenefittedfromthePOP-ALERTdashboard.

TheanswersillustratedinTable9alsorevealthatusingthedashboardhasimprovedtheawarenessandpreparednessofparticipants.Thedifferenceintheresultsisnotparticularlysignificantwhenlookingatthemostpopularanswers:water,cannedfood,flashlightandradioareitemspreferredbybothgroups.However,ahighproportionoftheexperimentalgroup’smembershavealsoselectedtheremainingitems,whilethecontrolgroupshavepracticallyignoredthem.Forexample,toiletries,mobilephone,clothesandcashareexamplesofimportantitemsmostlyignoredbythecontrolgroup.

Finally,inthequestionregardingsourcesofinformationtheresultsshowninTable10aremostlysimilarbetweenthetwogroups.BothgroupshaveidentifiedTVandradioasprimarysourcesofinformation,whichisinaccordancewiththerecommendedcourseofaction.Weshouldnotethatasmallproportionofparticipantshavespecificallyidentifiedthedashboardandemergencyactioncardsincludedthereinasasourceofinformation,whichisparticularlyencouragingsincetheexistenceofthesesourceswasonlyrecentlyrevealedtotheusers.Finally,themajorityoftheexperimentalgrouphavenotincludedemergencyservicesintheiranswers,whichisalsotherecommendedcourseof action since thiswouldunnecessarily increase theworkloadof the respectivecall emergencycommunicationcentresdealingwithurgentpost-disasterincidents.

Comparison with Preparedness Survey ResultsItisworthcomparingtheLisbontrialresultswithanon-lineemergencypreparednesssurveycirculatedamongEUcitizensinwhichover1600peopleacrosstheEUparticipated(Filippoupolitisetal,2015).

Table11illustratestheanswertoaquestionrelatedtoreasonsbehinddelayingtheevacuationinanemergencysituation.Aswecanobserve,ahighnumberofparticipants(32.7%)statedthattheywoulddelayevacuatinginordertocarefortheirpets.GoingbacktoTable4,wecanconfirmthatthemajorityofthecontrolgroup(51.72%)optedtotaketheirpetswiththemduringtheevolvingearthquakedisaster.Thisconfirmsthefactthattheanswerscomingfromsampleofpopulationusedinourfieldtrialisinaccordancewiththeresultscomingfromthebiggersamplethatparticipatedintheearliersurvey.

Anothercomparisonwecandrawbetweenthetwopopulationsamplesisrelatedtothesuppliesanditemstheyhaveallocatedforuseinthecaseofadisaster.Table12illustratestheresultsoftheEUsurvey.Wecanobservethatthereisasimilaritybetweenthetwosamples,specificallywithrespecttothemostpopularitems.Forexample,boththeparticipantsoftheEUsurveyandofthetrialhaveselectedwater,food,flashlightandradioasthemostpopularitemstoputintheiremergencykit.Intermsoftheleastpopularitems,thereareminorvariationsbetweenthetwogroups.However,itemssuchasclothingandfireblanketsareamongtheleastpopularchoicesforboththeEUandtheLisbon

Table 7. Responses to question on “If you are directly affected by the earthquake: What actions would you take when you are outdoors?”

If you are directly affected by the earthquake: What actions would you take when you are outdoors?


Gotoopenspace/awayfrombuildings 72.41% 58.33%

Seekfamily 0.00% 2.78%

Getinvehicle 0.00% 2.78%


30

Table 8. Responses to question on “If you are directly affected by the earthquake: What actions would you take when you are in a vehicle?”

If you are directly affected by the earthquake: What actions would you take when you are in a vehicle?


Stayinside 17.24% 38.89%

Stopcar 20.69% 27.78%

Drivetoopenarea 24.14% 27.78%

Leavecar 27.59% 2.78%

Table 9. Responses to question on “List the items you would expect to find in your emergency kit/bag”

List the items you would expect to find in your emergency kit/bag


Water 51.72% 61.11%

Cannedfood 51.72% 55.56%

Flashlight 44.83% 47.22%

Radio 41.38% 47.22%

FirstAidKit 27.59% 44.44%

Batteries 31.03% 41.67%

Medicine 20.69% 41.67%

Knife 10.34% 27.78%

Blanket 10.34% 25.00%

Documents 0.00% 25.00%

MobilePhone 0.00% 22.22%

Cash 3.45% 19.44%

Clothes 3.45% 19.44%

Toiletries 0.00% 19.44%

Bandages 13.79% 0.00%

FireExtinguisher 6.90% 5.56%

Matches/Lighter 6.90% 2.78%

Charger 0.00% 2.78%


31

Table 10. Responses to question on “If you are unsure of what actions should be taken, where would you find this information?”

If you are unsure of what actions should be taken, where would you find this information?


Radio 41.38% 47.22%

Internet 24.14% 22.22%

TV 24.14% 25.00%

Civilprotection 24.14% 5.56%

Call112 3.45% 19.44%

Fireservice 10.34% 0.00%

Authorities 10.34% 2.78%

Police 3.45% 5.56%

Smartphone 17.24% 2.78%

SocialMedia 10.34% 2.78%

Firstresponders 10.34% 2.78%

EmergencyContacts 0.00% 5.56%

Discussingwithothers 3.45% 2.78%

POP-ALERTplatform 0.00% 2.78%

Don’tknow 0.00% 2.78%

EmergencyActioncards 0.00% 2.78%

Relatives 0.00% 2.78%

Table 11. Responses to the EU survey question on “Would you consider delaying evacuating in an emergency situation for any of the following reasons?”

Would you consider delaying evacuating in an emergency situation for any of the following reasons?

Evacuatemypet(s) 32.7%

Gathermypersonalbelongings 33.7%

Gainclarityintheunfoldingevent 48.6%

Waitfordirectionsfromemergencymanagementagencies 40.2%

Personalmobilityissues 5.9%

Careforrelative 45.7%

Iwouldnotdelayevacuating 17.3%

Iwouldnotevacuate 1.2%

Other 2.9%


32

participants.TheseobservationsalsoreinforcethefactthatthecharacteristicsofourtrialssampleareinaccordancewiththoseobservedintheEUsurveyparticipants.

CONCLUSION

TheworkdescribedinthispaperisbasedonanumberofyearsofresearchbothintodistributedeLearningandcrisismanagementandresponse,andasaresultoffersaresearch-informedpracticalcontributiontotheexistingprovisionofservicesforpopulationawareness,trainingandpreparation.Itstillremainsthecasethatmuchofthetrainingonofferincrisisresponseiseitherpaper-based,orexpensivefieldeventsorsimulations,primarilytargetedatfirstrespondersintheemergencyservices.ThePandoraprojectsoughttoprovidearichmultimedia,immersivetrainingenvironmentforstrategiccrisismanagers(GoldCommanders)thatcouldofferrelativelylowcost,repeatableandself-configuredtrainingevents.ThePandora+environmentoffersaproduction-levelsystemtoprovidesuchtrainingacrossawidevarietyofdomains,onacommercialbasis.Linkingthisintotheotherresearchworkdescribedinthepaper,aunique,scalablemodelforthedesignoflearningmaterialstosupporttrainingincrisispreparationandresponsehasbeendemonstrated.ResearchworkoneLearning,particularlyfocusedontheuseofMOOCs,highlightsanumberofissuesofrelevancetoofferingamassivelyopenanddistributedtrainingcourseoncrisisresponseandpreparation.Inparticular,buildingabitesizemodelofdelivery,toreflectuserpreferenceforquickandshortengagement,acceptingusermodelsofengagementwhenevertheywishandforaslongastheywish,andrewardingengagementratherthancompletion.Intermsofeducatingthepopulation,particularencouragementcanbedrawnfromawillingnessexpressedbythepopulationtoprepare,butalackofknowledgeofhowto,orwheretofindtheinformation,isholdingbackthepreparednessofcitizens.

Onthebasisofthisbackgroundresearch,thePandora+environmenthasbeenrevisedtooffersupportforboththeexistingtrainer-ledexercisemodelforwhichitwasoriginallydesigned,andforawidelydistributed,webservicemodelallowingrelativelyrandomtrainee(membersofthepublic)engagement.ThesetrialshaveprovidedvaluablefeedbackonboththePOP-ALERTdashboardanduseofthePandora+system,bothofwhichwereextremelypositive,andnowprovidesthegreenlighttoprovidetrainingoncrisismanagementtothegeneralpubliconalargescale,withthepotentialtohaveasignificantimpactonthepreparednessoftheEUpopulationforacrisis.Theauthorsbelievethat,notonlyhavetheenhancementstothePandora+systemdemonstratedthecapacitytoprovideadistributed,online,immersive,multimedia,engagingenvironmentdesignedformasseducationofthepublictosupportthemindevelopingtheircrisispreparationandresponseskillsonthescaleofaMOOC,butthattheresultsofthetrialsandthesurveyhavepotentialtoinformcurrentthinkingoncrisispreparednesseducationforthepopulation.FutureenhancementstoPandora+includethereleaseofmoretrainingscenarios,furtherdevelopmentofthebehaviouralframeworktoautomatethestressmanagementofatraineeintheabsenceofatrainer,andtoprovideintelligentagentsasscenarioavatarstoaugmenttheimmersivecharacteristicsofascenario.

ACKNOwLEDGMENT

TheauthorswouldwishtoacknowledgethesupportoftheEuropeanCommissionforfundingthefollowingthreeprojects.

1.PandoraprojectundertheIST&SecurityCallofthe7thFramework-FP7-ICT-SEC-2007-1,GrantAgreementNo.225387

2.POP-ALERTprojectunderthesecuritythemeofthe7thFramework,callSEC-2013.4.1-5grantagreement608030.

3.dCCD-FLITEprojectundertheErasmusLifelongLearningprogramme,agreementnumber2013-4354/001-001,projectnumber539249-LLP-1-2013-1-SE-ERASMUS-EKA.

We would also wish to acknowledge and thank our partners in all three projects for theircontributionstothiswork.


33

Table 12. Responses to the EU survey question on “Which of the following supplies do you have in your home in case of a disaster?”

Which of the following supplies do you have in your home in case of a disaster?

Watercontainer 47.8%

Non-perishablefood 64.6%

Flashlight 81.1%

Batteryoperatedradio 32.4%

Batteries 74.3%

Firstaidkit 70.4%

Medication 71.2%

PhotocopiesofID 36.5%

Batteryoperatedmobilephone 54.2%

Mobilephonecharger 83.3%

Candles 79.1%

Whistle 31.1%

Matches/Lighter 84.7%

Fireblanket 21.7%

Protectiveclothing 36.7%

Noneoftheabove 1.7%

Other 2.9%


34

REFERENCES

Bacon,L.,MacKinnon,L.,Anderson,M.,Hansson,B.,Fox,A.,Cecowski,M.,...Stamatis,D.(2015).AddressingretentionandcompletioninMOOCs-astudent-centricdesignapproach.InE-LEARN 2015 - World Conference on E-Learning,Kona,Hawaii,October19-22.

Bacon,L.,MacKinnon,L.,Cesta,A.,&Cortellessa,G.(2012).DevelopingaSmartEnvironmentforCrisisManagementTraining.Journal of Ambient Intelligence and Humanized Computing,3(2).doi:.Retrievedfromhttp://www.springerlink.com/content/r586h5354923/?MUD=MP10.1007/s12652-012-0124-0

Bacon,L.,MacKinnon,L.,&Kananda,D.(2015).Real-timeDecisionSupportinOnlineTrainingEnvironments.InProceedings of the XIII LACCEI Conference, Santa Domingo July29-31.doi:10.18687/LACCEI2015.1.1.192

Bagrow, J., Wang, D., & Barabási, A.-L. (2011). Collective response of the human populations to largescale emergencies. Journal Plos One, 6(3), e17680. Retrieved from http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0017680#pone-0017680-. doi:10.1371/journal.pone.0017680PMID:21479206

Bernard,R.M.,RojodeRubalcava,B.,&St-Pierre,D.(2000).Collaborativeonlinedistancelearning:Issuesforfuturepracticeandresearch.Distance Education,21(2),260–269.doi:10.1080/0158791000210205

Dow,K.,&Cutter,S.L.(1998).Cryingwolf:Repeatresponsestoevacuationorders.Coastal Management,26(4),237–252.doi:10.1080/08920759809362356

Dow,K.,&Cutter,S.L.(2000).Publicordersandpersonalopinions:Householdstrategiesforhurricaneriskassessment.Environmental Hazards,2(4),143–155.doi:10.1016/S1464-2867(01)00014-6

Ellis,A.(2001).Student-centredcollaborativelearningviaface-to-faceandasynchronousonlinecommunication:What’sthedifference.InProceedings 18th ASCILITE Conference Melbourne.

Filippoupolitis,A.,MacKinnon,L.,&Bacon,L.(2015).ASurveyonEmergencyPreparednessofEUCitizens.InProceedings of the ISCRAM 2015 Conference,May24-27.

Klaiman,T.,Knorr,D.,Fitzgerald,S.,DeMara,P.,Thomas,C.,Heake,G.,&Hausman,A.(2010).Locatingand communicating with at-risk populations about emergency preparedness: The vulnerable populationsoutreachmodel.Disaster Medicine and Public Health Preparedness,4(3),246–251.doi:10.1001/dmp.2010.25PMID:21149222

Kop,R.,Fournier,H.,&SuiFaiMak,J.(2011).Apedagogyofabundanceorapedagogytosupporthumanbeings?Participantsupportonmassiveopenonlinecourses.International Journal of Research in Open and Distance Learning,12(7).Retrievedfromhttp://www.irrodl.org/index.php/irrodl/article/view/1041/2025

Lalo,A. (2000).Alerting thepopulation inemergencyplans:Examplesof localpublicpolicy inProvence.Journal of Hazardous Materials,78(1-3),281–301.doi:10.1016/S0304-3894(00)00227-2PMID:10978571

Legates,D.,&Biddle,M.(1999).Warning response and risk behaviour in the Oak Grove - Birmingham, Alabama Tornado of April 8, 1998.Boulder,CO:NaturalHazardsResearchandApplicationsInformationCentre,InstituteofBehaviouralScience,UniversityofColorado.

Leonard,G.S.,Johnston,D.M.,Paton,D.,Christianson,A.,Becker,J.,&Keys,H.(2008).Developingeffectivewarningsystems:OngoingresearchatRuapehuvolcano,NewZealand.Journal of Volcanology and Geothermal Research,172(3),199–215.doi:10.1016/j.jvolgeores.2007.12.008

Lindell,M.K.,&Perry,R.W.(2004).Communicating Environmental Risk in Multiethnic Communities.ThousandOaks,CA:SagePublications.

MacKinnon,L.,Bacon,E.,Cortellessa,G.,&Cesta,A.(2013).UsingEmotionalIntelligenceinTrainingCrisisManagers:ThePandoraApproach.The International Journal of Distance Education Technologies,11(2),66-95.Retrievedfromhttp://www.igi-global.com/article/using-emotional-intelligence-training-crisis/77841

Mileti,D.S.,&Peek,L.(2000).Thesocialpsychologyofpublicresponsetowarningsofanuclearpowerplant accident. Journal of Hazardous Materials, 75(2–3), 181–194. doi:10.1016/S0304-3894(00)00179-5PMID:10838242

http://www.springerlink.com/content/r586h5354923/?MUD=MP

http://dx.doi.org/10.18687/LACCEI2015.1.1.192

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0017680#pone-0017680-

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0017680#pone-0017680-

http://dx.doi.org/10.1371/journal.pone.0017680

http://www.ncbi.nlm.nih.gov/pubmed/21479206

http://dx.doi.org/10.1080/0158791000210205

http://dx.doi.org/10.1080/08920759809362356

http://dx.doi.org/10.1016/S1464-2867(01)00014-6

http://dx.doi.org/10.1001/dmp.2010.25


http://www.irrodl.org/index.php/irrodl/article/view/1041/2025

http://dx.doi.org/10.1016/S0304-3894(00)00227-2


http://dx.doi.org/10.1016/j.jvolgeores.2007.12.008

http://www.igi-global.com/article/using-emotional-intelligence-training-crisis/77841

http://dx.doi.org/10.1016/S0304-3894(00)00179-5



35

Paton,D.(2006).RiskCommunicationproject(C4)Warningsystems:Issuesandconsiderationforwarningthe public (Draft report). Retrieved from http://www.bushfirecrc.com/managed/resource/paton-bushfire-warnings_wcover_1.pdf

Perry,R.W.,Lindell,M.K.,&Greene,M.R.(1982).Crisiscommunications:Ethnicdifferentialsininterpretingandactingondisasterwarnings.Social Behaviour and Personality: An International Journal,10(1),97–104.doi:10.2224/sbp.1982.10.1.97

POP-ALERT.(2015).(Projectwebsite).Retrievedfromhttp://www.pop-alert.eu/

Quarantelli, E. (1994). Perceptions and reactions to emergency warnings of sudden hazards. Ekistics, 309,511–515.

Salas,E.,Driskell,J.,&Hughs,S.(1996).Thestudyofstressandhumanperformance.InJ.Driskell&E.Salas(Eds.),Stress and Human Performance.NewJersey:LawrenceErlbaumAssociates.

Seydlitz,R.,Spencer,J.W.,&Lundskow,G.(1994).MediaPresentationsofaHazardEventandthePublic’sResponse:AnEmpiricalExamination.International Journal of Mass Emergencies and Disasters,12,279–301.

Vihalemm,T.,Kiisel,M.,&Harro-Loit,H.(2012).Citizens’responsepatternstowarningmessages.Journal of Contingencies and Crisis Management,20(1),13–25.doi:10.1111/j.1468-5973.2011.00655.x

http://www.bushfirecrc.com/managed/resource/paton-bushfire-warnings_wcover_1.pdf

http://www.bushfirecrc.com/managed/resource/paton-bushfire-warnings_wcover_1.pdf

http://dx.doi.org/10.2224/sbp.1982.10.1.97

http://www.pop-alert.eu/

http://dx.doi.org/10.1111/j.1468-5973.2011.00655.x


36

Liz Bacon BSc, PhD, CEng, CSci, CITP, FBCS, PFHEA, MACM is a Deputy Pro-Vice-Chancellor at the University of Greenwich in London, with a University wide remit leading the development of technology enhanced learning. She was President of BCS, The Chartered Institute for IT, in the year 14-15 and is a past Chair of the BCS Academy of Computing, and the CPHC (Council of Professors and Heads of Computing) national committee. Liz is a Professor of Software Engineering with over a hundred publications and a Co-Director of the eCentre research group. She has been involved in several EU research projects, including being Principal Investigator and Project Coordinator for two EU FP7 projects in the recent past. She is an experienced systems designer and developer and her research interests include computing policy, smart systems, security, technology enhanced learning and crisis management.

Lachlan MacKinnon BSc, PhD, FBCS, CITP, CEng, MIEEE, MACM, MAACE is Professor of Computing Science (Strategic Development), and Head of the Department of Computing & Information Systems, in the Faculty of Architecture, Computing & Humanities, University of Greenwich, U.K. He is also Visiting Professor of Computer Science at RMIT University, Melbourne, Australia, and Visiting Professor of Games & Multimedia Technology at University College of SouthEast Norway, Kongsberg, Norway. Professor MacKinnon is a former Chair of the UK national Committee of the Council of Professors and Heads of Computing (CPHC), and Chair of the Education Committee of the BCS Academy of Computing. He is Chair of the Executive Committee of the British National Conference on Databases, and a member of the UK National Committee of the British Human Computer Interaction Group. His research interests are in computing policy, information and knowledge engineering, smart systems, games and creative technologies, eHealth and eLearning, and cybersecurity.

Avgoustinos Filippoupolitis is a senior lecturer in Disruptive Technologies, carrying out research in distributed systems and machine learning with applications to the Internet of Things, Emergency Preparedness for large scale disasters and Cyber-profiling. Avgoustinos was awarded his PhD on emergency simulation and decision support algorithms for smart buildings from Imperial College. His work was competitively selected for demonstration in the Home Office INSTINCT programme, aiming at reducing the risk of terrorist attacks in crowded urban areas. He has collaborated with MBDA Systems under an EPSRC Knowledge Transfer Secondment for the exploitation of his research on discrete event simulation and modelling for urban security. He is currently an independent expert in the evaluation of Horizon 2020 proposals.

David Kananda has a BSc (Hons) in Computing from the University of Greenwich and has worked on the Pandora+ project as a software developer for over 4 years. His specialist areas of expertise are in web service development, and cloud deployment, using technologies such as Java, JavaScript, Bootstrap, Angular, Docker and AWS inter alia. He also has expertise in database design and development using MySQL, MongoDB and NoSQL, deployment tools and testing, and was the designer of the decision branching mechanism within the Pandora+ system.