21

DETERMINANTS OF BRAND LOYALTY AND THE LINK BETWEEN BRAND LOYALTY AND PRICE TOLERANCE

EBRu TüMER KABADAYI* İncİ Aygün**

Gebze Institute of Technology Gebze Institute of Technology

ABSTRACT

The aim of this study is to investigate the integrated relationships between brand satisfaction, brand trust and brand affect through the brand loyalty development process and price tolerance as an outcome of the brand loyalty process. Data are collected based on a survey of 1085 students from various universities in Istanbul, Turkey. The findings indicate that brand affect and brand trust are key constructs in the brand loyalty formation process. Brand satisfaction is effective on brand loyalty indirectly through strengthening these constructs. Additionally, brand loyalty is a strong antecedent of consumers’ tolerance to price. The research findings will help provide clarification on how brand loyalty comes about and brings about price tolerance.

Key words: brand loyalty, brand trust, brand satisfaction, brand affect and price tolerance.

MARKA BAĞLILIĞINI BELİRLEYEN FAKTÖRLER VE MARKA BAĞLILIĞI İLE FİYAT TOLERANSININ İLİŞKİSİ

ÖZET

Bu çalışmanın amacı, marka bağlılığı geliştirme süreci içinde marka memnuniyeti, markaya duyulan güven ve markanın oluşturduğu duygular arasında iç içe geçmiş ilişkilerin ve marka bağlılığı sürecinin çıktısı olarak fiyat toleransın incelenmesidir. Veriler İstanbul’daki çeşitli üniversitelerde okuyan 1085 öğrenci üzerinde yapılan anket çalışması ile toplanmıştır. Bulgular, markanın oluşturduğu duyguların ve markaya duyulan güveninin marka bağlılığı oluşturma sürecindeki temel faktörler olduğunu göstermektedir. Marka memnuniyeti ise, bu temel faktörleri güçlendirerek marka bağlılığı üzerinde dolaylı olarak etkili olmaktadır. Ayrıca, marka bağlılığının müşterilerin fiyata olan toleransının güçlü bir öncülü olduğu ortaya çıkmıştır. Araştırmanın bulguları marka bağlılığının nasıl oluştuğunu ve fiyat toleransını nasıl meydana getirdiğini anlamamıza yardımcı olacaktır.

Anahtar kelimeler: marka bağlılığı, markaya duyulan güven, marka memnuniyeti, markanın oluşturduğu duygular

ve fiyat toleransı.

* EbruTümerKabadayıisanAssistantProfessorintheFacultyofBusinessAdministrationattheGebzeInstituteofTechnology,ÇayırovaFabrikaYolu,No:101,41400,Gebze,Kocaeli,Turkey.E-mail:tumer@gyte.edu.tr

** İnciAygünisaResearchAssistantintheFacultyofBusinessAdministrationattheGebzeInstituteofTechnology,ÇayırovaFabrikaYolu,No:101,41400,Gebze,Kocaeli,Turkey.E-mail:iaygun@gyte.edu.tr

Boğaziçi Journal Vol. 21, no. 1-2 (2007), pp. 21-35.

22

Thelong-termsuccessofabrandisnotbasedonthenumberofconsumerswhobuyitonce,butonthenumberofconsumerswhobecomeregularbuyersofthebrand(Odinetal.,2001).Thisstatementsimplydemystifiesfirms’strictaimstoseek,manageandincreasebrandloyaltytogainadvantageoratleasttosurviveespeciallyinintenselycompetitive,uncertain,andhardlypredictablemarkets.

Aaker(1996)statesthatabrand’svaluetoafirmislargelycreatedbythecustomerloyaltyitcommands.Theliteratureonbrandmanagementprovidesalargeamountofevidenceconcerningpositiveoutputsrelatedtobrandloyalty,suchashighprofitabilityandcompetitiveadvantageforfirms.Theincreasedprofitfromloyaltycomesfrompricepremiums,greatermarketshare(ChaudhuriandHolbrook,2001)andreducedmarketing-operationalcosts(BowenandChen,2001).AsChaudhuriandHolbrook(2001)proposedandconfirmedthatbrandshigherinpurchaseloyaltyarealsohigherinmarketsharebecauseofgreaterlevelsofrepeatpurchasebybrandusersandthatgreaterattitudinalloyaltyshouldleadtogreaterwillingnesstosacrificebypayingapremiumpriceforavaluedbrand.Brandloyaltymayalsohelptoincreasemarketshareindirectlybyprovidingpositivewordofmouth,whichisoneofthemostpowerfulsourcesinpersuasion(Selnes,1993).Becausefindingnewcustomersanddoingbusinesswiththemtakestime,effortandmoney(MittalandLassar,1998)brandloyaltyreducesmarketingcostssinceloyalconsumersalreadyknowtheproductandrequirelessinformation.BallesterandAlleman(2001)indicateotheradvantagesbystatingthatbrandloyaltyprovidesacompetitiveadvantagebygeneratingasubstantialentrybarrier tocompetitors,an increase in thefirm’sability torespondtocompetitivethreats,andacustomerbaselesssensitivetothemarketingeffortsofcompetitorsinhighlycompetitivemarkets.Inaddition,brandloyaltyhasbeenshowntobeassociatedwithhigherratesofreturnoninvestmentthroughincreasesinmarketshare(GounarisandStathakopoulos,2004).

Thebenefitsofbrandloyaltymentionedabove,hasmadeitapopularissueforfirmsandmarketingresearchinthelastthreedecades.Itisoneofthemainpurposesofthisstudytoexplore theprocessthroughwhichbrandloyaltyisformed.Forthispurpose,theeffectsofbrandsatisfaction,brandtrustandbrandaffectareinvestigated.

Anotherpurposeofthisstudyistoinvestigatepricetoleranceasanoutcomeofbrandloyalty.Theconstructs representing consumers’ actual or intended reactions to different price levels such aspricetolerance,brandacceptanceandpricepremiumsareincreasinglystudiedinconsumerbehaviorliterature.Aaker’s(1996)descriptionofpricepremiumasareasonablesummaryofthestrengthofthebrandhelptounderstandtheimportanceoftheseconstructsinbranddomain.Pricetolerancewasincludedinourresearchmodelasaconstructrepresentingconsumer’sdifferingreactionstodifferentpricelevels.Althoughpricetoleranceismoreoftenproposedtobedirectlyrelatedtosatisfactioninpreviousstudies(e.g.,Anderson,1996;Herrmannetal.,2004),inthecurrentstudyitisinvestigatedasaconsequentofbrandloyalty.

Althoughvarious antecedents and consequentsofbrand loyaltyhavebeen separately evaluated innumerousstudies thatare formedondifferent relationshipnetworks, there isstilla lackofaclearunderstanding about the roles of these constructs. For extended and clearer knowledge about thebrandloyaltyconcepttheinteractionofantecedentsandconsequentsofbrandloyaltyneededtobeinvestigated.Forthispurpose,amodelwasdevelopedusingthestructuralequationmodelingapproach,which is adequate for evaluating an integratedmodel. Themodel contained the key determiningconstructsthataremostcommonlymentionedinthebrandloyaltyliteraturesuchasbrandsatisfaction,brandtrustandbrandaffectandpricetoleranceasapositiveoutcome.

23

Thestudybeginswithdescribingtheliterarybackgroundofthismodel.Next,themodelistestedandtheresultsaresummarizedanddiscussed.Finally,suggestionsforfutureresearcharegiven.

FRAMEWORKANDTHEHYPOTHESES

Brand Loyalty

Brandloyaltyhasbeeninvestigatedforalongtimebyresearchersandthereareamplestudiesonthisconcept.However,duetothedifferentperspectivesintheliterature,therearediversedefinitionsofbrandloyalty.In1973,JacobyandKynerpointedouttheproblemofthedefinitionsandoperationalizationoftheconstruct,statingthatsomanydefinitionsmadeitdifficultandhazardoustocompare,synthesizeandaccumulatefindings.Itappearsthatthecomplexityoftheconstructofloyaltystillpreventstheformulationofauniquedefinitionintherecentliterature.ThieleandMackay(2001)statethatinthemarketingliteraturetheterm“loyalty”hasbeenusedinterchangeablywithitsoperationaldefinitionto refer to repeatpurchase,preference,commitment, retentionandallegiance.These termsare theoutputsofdefinitionsthatarerelatedtothedifferentaspectsofbrandloyalty.

QuesterandLim(2003) state that the literature shows twoalternativeapproaches to theconstructofbrandloyalty.Thefirstoneisconcernedwith“aconsistentpurchasebehaviorofaspecificbrandovertime.”Thisisthebehavioralapproachtobrandloyaltyandithasbeenwidelyusedtodefinetheconstruct.Thesecondonereliesonafavorableattitudetowardsabrand.

Neitherapproachseemstobesufficientbyitselftodefinebrandloyaltycompletelyasbothignoreeitherthepsychologicalorbehavioraldynamicsoftheconcept.Oneproblemwiththebehavioralapproachisthatrepeatpurchasesarenotalwaystheresultofapsychologicalcommitmenttothebrand(BowenandChen,2001).Especiallyforlowinvolvementproducts,repeatpurchasingofthesamebrandmaybeahabitualactiontoavoidspendingthetimeandeffortnecessaryforevaluatingalternativebrands.The attitudinal approach,which includes emotional andpsychological attachment to abrand, also

24

hassomeproblemsas theremaybe reasons thatprevent theconsumer frompurchasing thebrandhe/shelovesorhasanattachment.QuesterandLim(2003)indicatethattruebrandloyaltyimpliesacommitmenttoaspecificbrandandgoesbeyondrepetitivebehavior.Inthisstudy,consistentwiththisidea,theconceptofbrandloyaltyincludesbehavioralintentiontobuythesamebrandandrecommendittoothers.Toencourageotherstobuywithstrongrecommendationemphasizestheattitudinalaspectofbrandloyaltyandthepurchaseintentionemphasizesthebehavioralaspect.

Inthemarketingliterature,therearenumerousstudiesthataimedtoexplorethesourcesandoutputsofbrandloyalty.Ashasbeenstatedbefore,theinteractionofbrandsatisfaction,trustandaffectthroughgeneratingbrandloyaltyandpricetoleranceasapositiveoutcomeofitwillbeexaminedinthisstudy.TheproposedmodelispresentedinFigure1.

Brand Satisfaction

Satisfactionisdefinedaspleasurablefulfillment,whichmeanstheconsumersensesthatconsumptionfulfillssomeneed,desire,goalorsoforthandthatthisfulfillmentispleasurable(Oliver,1999).Itis the resultofacognitiveandaffectiveevaluationwhere somecomparisonstandard iscomparedto actually perceived performance (Homburg and Giering, 2001). When the subject is a brand,the evaluation output appears as brand satisfaction versus brand dissatisfaction. Consistent withsatisfactiondefinitionsabove, in thisstudy,brandsatisfaction isconceptualizedas theoutcomeofthesubjectiveevaluationthatthechosenalternativebrandmeetsorexceedsexpectations(LouandLee,1999).Satisfactionisregardedasafundamentaldeterminantoflong-termconsumerbehavior(Ranaweera and Prabhu, 2003), thereby, as one of the key global constructs predicting consumerbehavior(GarbarinoandJohnson,1999).

Althoughitisknownthatbrandsatisfactionhasaroleindevelopingbrandloyalty,thecomplexityofthelinkbetweenthesetwoconstructsmakesithardtogetafullunderstanding.Becauseofthefactthatonlyarelativelysmallportionofsatisfiedcustomersreturnstothesameserviceorproduct(WhiteandYu,2005)andbecausemerelysatisfyingcustomersthathavethefreedomtomakechoicesisnotenoughtokeepthemloyal(Oliver,1999),brandsatisfactionislikelytoneedsomeotherconstructsasmediatingactorstobeeffectiveonbrandloyalty.Inthisstudy,itisproposedthatbrandsatisfactionhasanindirecteffectonbrandloyaltythroughstrengtheningbrandtrustandbrandaffect.

Consumersbuyproducts/services toobtaincertainphysical and/orpsychologicalbenefits.So tobewillingtostayinalong-termrelationshipwiththebrandtheyareverylikelytoneedtobelievethatthebrandwillconstantlyprovidethesamebenefitsandsatisfactionwitheveryproductwiththesamebrandname.Sobrandtrustisneededasamediatingactorbetweenbrandsatisfactionandbrandloyalty.

Thesatisfaction trustrelationshipcanbeproposedbasedonthecharacteristicsanddevelopmentprocessesof theseconstructs.Because trust requires anassessmentof theotherparty’s credibilityandbenevolence,onepartymusthaveinformationabouttheotherparty’spastbehaviorandpromises(DoneyandCannon,1997);thatis,consumptionexperienceisthemostrelevantandimportantsourceofbrandtrustbecauseitgeneratesassociations,thoughtsandinferencesthataremoreself-relevantandheldwithcertainty(BalesterandAleman,2005).Satisfaction,asmentionedabove,isapost-choiceevaluativejudgmentofaspecifictransaction(Selnes,1993)andcanprovidethenecessaryinformationforthebrandtrustassessment.LouandLee(1999)alsostatesatisfactionwithpastoutcomesindicatesequity in the exchange. This increases the perception of the exchange partner’s benevolence and

25

credibility,becausewhenaconsumerissatisfiedwithabrandafterusingit,thissituationissimilartopromisefulfillment.Sincethebrandhaskeptitspromise,theconsumerisliabletotrustitmore.

Hypothesis 1: Brand satisfaction is positively related to brand trust.

Furthermore,thebrandsatisfaction-brandloyaltyrelationshipissuggestedalsotohavebrandaffectas anothermediating actor.Brand satisfaction affects brand loyalty indirectlyby evokingpositivefeelingsaboutthebrandbecauseofthebenefitsitprovides.

Hypothesis 2: Brand satisfaction is positively related to brand affect

Thissuggestedrelationshipmayalsomeantheindirecteffectofsatisfactiononbrandtrustbecauseofthesuggestedeffectofbrandaffectonbrandtrustthatwillbediscussedbelow.

Brand Trust

MorganandHunt(1994)conceptualizetrustasexistingwhenonepartyhasconfidenceinanexchangepartner’sreliabilityandintegrity.Probablybecauseitisgenerallyviewedasanessentialingredientforsuccessfulrelationships,asstatedinGarbarinoandJohnson(1999),trustisoneofthemostfrequentlymentionedfactorsintherelationalmarketingliterature(MorganandHunt,1994;DoneyandCannon,1997; Selnes, 1998; Lau and Lee, 1999; Chaudhuri and Holbrook, 2001; Ballester and Aleman,2005).

In thebrand-consumer relationship, trust shouldalsobeacrucial factor. It shouldbe important tobuildatrustworthyimageforthebrandbecauseinmostcasesconsumersseemtomeetthebrandofaproduct/servicebeforemeetingtheproducer,salesperson,etc.BallesterandAleman(2005)defineatrustworthybrandasonethatconsistentlykeepsitspromiseofvaluetoconsumersthroughthewaytheproductisdeveloped,produced,sold,servicedandadvertised,eveninbadtimeswhensomekindofbrandcrisisarises.Followingthedefinitions,itisnothardtorelatebrandtrusttobrandloyalty,logically.Previousstudiesstrengthenthisexpectation.

Morgan and Hunt (1994) explain the trustcommitment link by pointing out that relationshipscharacterized by trust are so highly valued that partieswill desire to commit themselves to suchrelationships.DoneyandCannon(1997)alsostatethatthehighleveloftrustcharacteristicofrelationalexchangeenablespartiestofocusonthelong-termbenefitsoftherelationship.

Chaudhuri and Holbrook (2001) suggest that brand trust leads to brand loyalty or commitment,becausetrustcreatesanexchangerelationshipthatoneishighlyinvolved.Indeed,commitmenthasbeendefinedasanenduringdesiretomaintainavaluedrelationship.Thus,loyaltyorcommitmentunderlies theongoingprocess ofmaintaining a valued relationship that has been createdby trust.Inotherwords, trustandcommitmentshouldbeassociatedbecause trust is important inrelationalexchangesandcommitmentisalsoreservedforsuchvaluedrelationships.

Inthiscontext,itcanbeproposedthatatrustedbrandshouldbepurchasedmoreoftenandshouldevokeahigherdegreeofattitudinalcommitment.Iftheconsumertruststhebrandhe/shewillbemorelikelytobuildalong-termrelationshipwiththebrandthatincludespositivebehavioralintentionandattitudinalloyalty.

26

Hypothesis 3: Brand trust is positively related to brand loyalty

Brand Affect

According toWestbrook (1987), affect is to compromise a class of mental phenomena uniquelycharacterized by a consciously experienced, subjective feeling state, commonly accompanyingemotionsandmoods.Inthesamestudy,Westbrook(1987)pointsoutthatadvancesinsocialcognition,cognitivepsychologyandsocialpsychologysuggestthataffectiveprocessesmayconstitutenotonlyapowerfulsourceofhumanmotivation,butalsoinfluenceinformationprocessingandchoice.Itappearsthattheaffectiveprocessmaybeadeterminantofconsumerbehaviorinthelongterm.Inthisstudy,affectiveresponsesrelatedtousageofabrandareproposedasantecedentsofbrandtrustandbrandloyalty.FollowingChaudhuriandHolbrook(2001),brandaffectisdefinedasabrand’spotentialtoelicitapositiveemotionalresponseintheaverageconsumerasaresultofitsuse.

Asmentionedabove, thebrandaffectissupposedtobeeffectiveonbrandloyaltybothinadirectandanindirectway.Indirectly,positivebrandaffectwillbeeffectiveonbuildingbrandloyaltybyincreasingtrustfeelingforthatbrand.Asaffectiveresponseshavebeenshowntobeabletoinfluencecognitiveprocessessuchasevaluation,recallandjudgment(Nyer,1997),consumers’affectiveresponsetoabrandusageislikelytobeeffectiveontheformationofbrandtrust.Soitcanbeproposedthathavingpositiveconsumptionelicitedaffectiveresponserelatedtobrandmayenableattributingatrustworthyimagetothatbrand.

Hypothesis 4: Brand affect is positively related to brand trust

Inaddition,brandaffectcanbeproposedasbeingaffectiveonloyaltydirectlybecauseloyaltywillbegreaterundertheconditionofmorepositiveemotionalmoodoraffect.Thatis,brandsthatmakeconsumers happy or joyful or affectionate should prompt greater purchase and attitudinal loyalty(ChaudhuriandHolbrook,2001).Inotherwords,positiveemotionalmoodoraffectaboutabrandwillencourageloyalty.

Hypothesis 5: Brand affect is positively related to brand loyalty

Price Tolerance

Pricetolerancecanbedescribedasthereactionofcustomerstothepriceincreaseofaspecificproductandisconstructedasapricespanwithintheboundariesofwhichtheconsumerdoesnotchangehis/herbuyingbehavior.Inthisway,thetoleratedpricerangestretchesfromtheactualpricepaidbytheconsumer to themaximumprice thataconsumer iswilling topay for theproduct (Hermanetal.,2004).Definitionsoftheconstructsabovearelikelytohelpusbuildthelogicallinkbetweenbrandloyaltyandpricetolerance.Butitisalsopossibletofindevidenceforthislinkinsomestudiesthatinvestigateconsumers’priceacceptance.Forinstance,Huberetal., (2000)investigatetherelationshipbetweencustomersatisfactionandpriceacceptancethatwasconceptualizedasthepotentialbuyers’willingnesstopurchaseasafunctionofvariousprices.Asseeninthisaspect,pricetoleranceislikelyto represent the latitudeofbrandacceptance. Latitudeofbrandacceptance is a concept that alsowasusedinKalyanaramandLittle(1994),whomadeanempiricalanalysisofthelatitudeofpriceacceptanceinconsumerpackagedgoods.Theysuggestthatconsumerswhoareontheaveragemore

27

brandloyalinagivenproductcategorywouldbelikelytohaveawiderlatitudeofpriceacceptanceforthatbrandbecausebrandloyaltywouldkeeptheconsumermorefocusedonthebenefitsofthebrandand less focused on price.Chaudhuri andHolbrook (2001) found a positive relationship betweenfavorablebrandattitudeandbeingmorewillingtopaypremiumpricesforabrand.

Hypothesis 6: Brand loyalty is positively related to price tolerance

METHOD

Data collection

Althoughitmayreducetheexternalvalidityofthestudybylimitingthegeneraluseofthefindings,datawerecollectedwithrespecttoonlyoneproductcategory,namelyjeans.Studyingonlyoneproductcategorymayhighlighttheprocessthroughwhichbrandloyaltyformsandleadstopricetolerancemoreclearlybyeliminatingthedifferencesinresponsepatternsduetodifferentreferencepoints.Jeanswerechosenastheproductcategorybecausemostpeoplearefamiliarwiththem,theyarepurchasedbydifferentsegmentsofthepopulationandtherearemanyalternativebrandswithdifferentquality,price,reputation,etc.Also,thisproductmayaddressbothutilitarianandhedonicneedsthatmayhelptosatisfyconstructs’differentaspectsifavailable.

DatawerecollectedfromasamplecomposedofstudentsfromnineuniversitieslocatedinIstanbul,Turkey.Theuniversitieswerechosenconsideringtheirtypeandlocation.TwoprivateandtwopublicuniversitiesontheEuropeansideofIstanbulandthreepublicandtwoprivateuniversitiesonAsiansidewerechosen.Atotalof1117questionnairesweredeliveredtostudentsonthecampusesoftheseuniversities.

Respondentswereasked toname the jeanbrand theyhadboughtmost recently andanswered thequestions thinking about that brand.After the elimination of some receivedquestionnaires due toexcessiveamountsofmissingdata,1085wereusedforanalysisthatweresubstantiallycompleteandresultedinavalidresponserateof82.74%.

Measurement

BrandsatisfactionwasmeasuredusingascaledevelopedbyLauandLee(1999).Ofthesevenitemsinthescale,fiveitemswereused,ofwhichthreewerereversed.Itemsaimed tomeasureconsumers’cognitivereactionsrelatedtoproductusage.TheoperationalizationofbrandaffectinvolvedadaptingChaudhuri andHolbrook’s (2001) three-itemscale,whichmeasures consumer’spost-consumptionpositiveaffectiveresponses.ChaudhuriandHolbrook’s(2001)three-itemscalewassupplementedbyanadditionalitemfromthestudyofLouandLee(1999),andtheresultingfour-itemscalewasusedtomeasurebrandtrust.Itinvolveditemsthataimedtoexploredirectlywhethertheconsumerfeltsecurityaboutthebrand.Eightitemswereusedtomeasurebrandloyalty,composedofthebehavioralandattitudinalaspectsoftheconstruct.ItemsweredrawnfromLouandLee’sscale(1999).Finally,following Ballester and Aleman (2001), price tolerance was measured by asking respondents iftheywouldbewillingtopaymoreforthebrandcomparedtootherbrands.MeasurementitemsarerepresentedinTable2.

28

All variables were measured based on a five-point Likert scale ranging from 1=strongly disagreeto 5=strongly agree. The questionnaire was pre-tested with 45 university students, and the finalquestionnairewasdevelopedwithrevisionofsomeitemsforclarityofwording.Toavoidresponsebias,thequestionnairewasdesignedintwoformsincludingthesamescaleitemsbutindifferentorder.

DATAANALYSISANDRESEARCHRESULTS

Table1givessummaryinformationaboutthesampleprofile.Ascanbeseen,thesampleishighlyconcentrated(74.9%)inthe20-24agerange.Mostoftheparticipantswereunemployedrespondentsfrompublicuniversitieswithincomesinthe1000-2500YTLincomerange.

Table 1Sample Statistics

Monthlypersonalincome*

Monthlyhouseholdincome

Percent Percent Percent Percent

Gender Employmentstatus Lessthan300YTL

22.7 1.3

Male 40.4 Employed (fullorpart-time)

15 300-600YTL

36.7 9.3

Female 59.6 Unemployed 85 601-999YTL

15.1 13.8

Age 1000-1499YTL

12.6 22.3

From15to19

19 Universitytype 1500-2499YTL

10.9 17.5

From20to24

74.9 Publicuniversity 68.1 2500-3499YTL

10.1

From25to29

5.9 Privateuniversity (withscholarship)

9.1 3500-5000YTL

1.7 9.1

30andolder

0.4 Privateuniversity (noscholarship)

22.8 Morethan5000YTL

0.8 16.6

*Monthlypersonalincomerepresentsincomeofemployedrespondents

Validity and Reliability of Scale

Initiallyanexploratoryfactoranalysiswascarriedoutemployingprincipalcomponentanalysisandrotatedusingvarimaxrotation.Atotaloftwoitemswereeliminated,onefromthebrandsatisfactionscaleandonefromthebrandloyaltyscale,ofwhichtheloadingonrespectivefactorwasbelow0.5.Thenallmeasureswereanalyzedforvalidityandreliabilitythroughconfirmatoryfactoranalysis.

29

Apurifiedfullmeasurementmodelwasproposedwith four reflectivemultiple itemfactors:brandsatisfaction, brand trust, brand affect, brand loyalty (with attitudinal and behavioral dimensions).The single item factor “price tolerance” was added to the model, setting its measurement errorat 10%variance of respective factor. This fullmeasurementmodelwas tested inAMOS5 usingthe maximum likelihood estimation techniques. As presented in Table 2, Goodness of Fit Index (GFI) = 0.91; Comparative Fit Index (CFI) = 0.92; Root Mean Square Error of Approximation(RMSEA) = 0.07 were in acceptable ranges although the chi-square statistic was very large (χ2

(141)=1027.032p<0.01),probablyduetothelargesamplesize.

To assess the reliability, the Cronbach’s alpha coefficient and composite reliability scores werecalculated.AspresentedinTable2compositereliabilityscoreswerewelloverthecriticallevel0.6(Fornell andLarcker,1981)and theCronbach’salphacoefficientswereover0.7 (Nunnaly,1978)indicatingreliabilityof themeasurementmodel.Estimatesofpathsfromindividual itemsto latentfactorsarealsoshowinTable2andtheyareallsignificant(p<0.01),providingevidenceforconvergentvalidity.

Discriminant validity was evaluated based on Fornell and Larcker’s (1981) criterion. The sharedvariance between pairs of latent factors in the structuralmeasurementmodelwas comparedwithaveragevarianceextractedthatwascalculatedforeachcomponentofpairs.Itwasfoundthataveragevarianceextractedwasgreater,providingevidencefordiscriminantvalidity.Itcanbeconcludedonthebasisofthereliabilityandvalidityanalysisthatscalesfortheconstructsappeartohaveacceptablemeasurementquality.Themeans,standarddeviations(SD)andinter-correlationsamongvariablesarerepresentedinTable3.Inthistable,theupperpartrepresentsthecorrelationsamonglatentfactors,andthepartbelowrepresentsthecorrelationsamongaggregatedmeasures.

Tests of the Hypotheses

The proposed research model, presented in Figure 1, was tested using the maximum likelihoodmethodinAMOS5.Goodnessoffitstatisticswereattheacceptablelevels(GFI=0.90;CFI=0.91;RMSEA=0.07),supportingtheoverallfitofproposedmodeltoourdata,althoughthechi-squarestatisticwasχ2

(141)=1028.974p<0.01.Theunexpectedlylargechi-squarevaluecanbeattributedtotherelativelylargesamplesizeinourresearch.Alsothestandardizedestimates,showninTable4,areallsignificantatp<0.01indicatingthatallofthehypothesizedlinksaresupported.Links,theirstandardizedestimates,andsignificancelevelsarepresentedinTable4.

TheresultsforH1and H2 supporttheconclusionthatconsumers’satisfactionwithabrandiseffectiveonbothdevelopingtrustandpositiveaffecttothatbrand.AlsoH4,suggestingthepositiverelationshipbetweenbrandaffectandbrandtrust,wassupported,meaningthatconsumers’positivefeelingaboutusingthatbrandiseffectiveonbuildingtrusttowardsthatbrand.

Continuingwiththeantecedentsofbrandloyalty,theresearchresultspresentempiricalsupportforthepositiveeffectofbrand trustandbrandaffectondevelopingbrand loyaltyashypothesizedbyH3andH5. Thestandardizedestimatesrevealthatbrandloyaltydependsonbrandtrustalittlemorethanbrandaffect(0.497and0.428;p<0.01).Finally,consideringtheconsequentsofbrandloyalty,resultsconfirmthatbrandloyaltyhasaninfluenceonthepricetoleranceofconsumers.Thus,thelasthypothesisH6wasalsosupported.

30

Table 2Measurement Items

Measurement item Estimates Alpha SCR AVE

Brand Satisfaction (BS) .80 .80 .50

ThisbrandhasnotworkedoutaswellasI

thoughtitwould.

.695**

Iamsatisfiedwithmydecisiontobuythis

brand

.574**

IamnothappythatIboughtthisbrand .796**

Ifeelbadaboutmydecisiontobuythisbrand .737**

Brand Trust (BT) .88 .88 .64

Itrustthisbrand .830**

Irelyonhisbrand .806**

Thisisanhonestbrand .747**

IfeelsecurewhenIbuythisbrandbecauseI

knowthatwillneverletmedown

.825**

Brand Affect (BA) .90 .90 .75

IfeelgoodwhenIusethisbrand .818**

Thisbrandmakesmehappy .883**

Thisbrandgivesmepleasure .897**

Brand Loyalty (BL) .76 .92 .85

Behavioral(intentional) .961** .74

Idointendtokeepbuyingthisbrand .689**

Ifthisbrandisnotavailableinthestorewhen

Ineedit,Iwillbuyitanothertime

.654**

Ifthisbrandisnotavailableinthestorewhen

Ineedit,Iwillbuyitsomewhereelse

.512**

IfanotherbrandishavingasaleIwillbuythe

otherbrandinsteadofthisone*

.689**

Attitudinal(emotional) .893** .75

Ifsomeonemakesanegativecommentabout

thisbrandIwoulddefendit

.742**

Ioftentellmyfriendshowgoodthisbrandis .653**

Iwouldnotrecommendthisbrandto

someonewhocannotdecidewhichbrandto

buyinthisproductclass*

.713**

Price Tolerance (PT) Not

available

Not

available

Not

available

Iwouldbewillingtopaymoreforthisbrand .950**

χ2 (141)=1027.032p<0.01;GFI=0.91;CFI=0.92;RMSEA=0.07

**p<0.01

31

Table 3Descriptive Statistics

Mean Std.Deviation 1 2 3 4 5

1.BrandSatisfaction 4.0645 .7321 1 .47** .29** .36** .25**

2.BrandTrust 3.4126 .8775 .41** 1 .65** .78** .55**

3.BrandAffect 3.3015 1.0532 .29** .58** 1 .75** .56**

4.BrandLoyalty 2.8890 .8127 .28** .63** .61** 1 .74**

5.PriceTolerance 2.79 1.32 .24** .49** .51** .61** 1

**p<0.01

Table 4The Standardized Path Estimates and Fit Statistics

Hypotheses Link Estimate S.E. C.R. P

H1 BrandSatisfaction BrandTrust .302 .039 9,611 .000

H2 BrandSatisfaction BrandAffect .294 .052 8,270 .000

H3 BrandTrust BrandLoyalty .497 .035 12,629 .000

H4 BrandAffect BrandLoyalty .428 .028 11,466 .000

H5 BrandAffect BrandTrust .556 .026 17,973 .000

H6 BrandLoyalty BrandTolerance .732 .061 19,604 .000

Chi-square=1028.974Degreesoffreedom=145Probabilitylevel=0.000Goodnessoffitindex(GFI)=0.90Comparativefitindex(CFI)=0..91Rootmeansquareerrorofapproximation(RMSEA)=0.07

Table 5Results of the Hypothesis Tests

Hypotheses Link Results

H1 BrandSatisfaction BrandTrust Supported

H2 BrandSatisfaction BrandAffect Supported

H3 BrandTrust BrandLoyalty Supported

H4 BrandAffect BrandLoyalty Supported

H5 BrandAffect BrandTrust Supported

H6 BrandLoyalty BrandTolerance Supported

32

DISCUSSION

Thisstudy investigates therolesofbrandsatisfaction,affectand trustas interactedantecedentsofbrand loyaltyand the linkbetweenbrand loyaltyandprice tolerance.Theanalysis resultsprovidesupportforalllinkagesthatarehypothesizedintheresearchmodel.

Intheliterature,satisfactionisfrequentlyinvestigatedinvariouspositionsasoneofthemainantecedentsofloyalty.Inthepresentstudy,itwassupposedthatsatisfactionwouldbeinanindirectrelationshipwithbrandloyalty.Intheresearchmodelproposingthatconsumer’ssatisfactionwouldrequiresomemediatingactorstoaffectbrandloyalty,thesatisfactionbrandloyaltydirectrelationshipwasnotallowed. The acceptable fit statistics confirmed this overall proposition. Moreover, modificationindiceswerecheckedand itwasfound that therewasnoother linkneeded.This findingprovidesadditionalsupportforthesuggestionabouttheindirectrelationshipbetweenbrandsatisfactionandbrandloyalty.Inthisindirectrelationship,themediatingroleofbrandtrustandbrandaffectcanbeseenclearly.TheresultsconfirmedH1thatsupposedthelinkofbrandsatisfactionbrandtrust(estimate1

=0.30p<0.01).Brandsatisfactionwasalsofoundtohaveaneffectonbrandaffect(estimate2 =0.29p<0.01)asstatedinH2.Itseemsthatconsumer’ssatisfactionlevelwaseffectiveatasimilarlevelbothonbeing formedofpositive feelingsaboutusing thebrandand the trustworthy imageof thebrand.Intheresearchmodelitwasalsoproposedthatconsumerswholikeusingabrandwouldbemoreinclinedtofindapositiveresultattheendoftheevaluationaboutthetrustworthinessofthatbrand.AnalysisresultsconfirmedH4,whichproposedthatbrandaffectwouldbeeffectiveonbrandtrust.Inotherwords,consumersareinfluencedbytheirfeelingsofusingthatbrandinattributingatrustworthyimagetothatbrand(estimate4=0.56p<0.01).Inthisrelationshipnetwork,halfof thevarianceinbrandtrustisexplainedbybrandsatisfactionandbrandaffect.Butthebrandaffectfactorwasexplainedintheratioof9%.Itseemsthatthereisanotherfactor(s)effectingconsumer’sfeelingsaboutbrandthatwasnotproposedintheresearchmodel.

Intheresearchmodel,brandtrustandbrandaffectwereproposedtobeantecedentsofbrandloyalty.ItwasnotunexpectedtoseeadirectandquitestrongeffectofbrandtrustonbrandloyaltyasH3 (estimate3=0.50p<0.01)becausetheimportantroleoftrustasadirectandalsoasamediatingfactorwasmentionedbefore in thebrandand relationship literature (LauandLee,1999;ChaudhuriandHolbrook,2001;BallesterandAleman,2005,e.g.)andsupportedbyempiricalevidence.Inthisstudy,additionallyanotherkeyvariable,brandaffectwasproposedtobeeffectiveonbrandloyaltydirectly.The roleofbrandaffect in theprocess inwhichbrand loyaltycomesabout isnearlyas strongasbrandtrusthavingasignificantdirecteffectasstatedinH5(estimate5= 0.43p<0.01) andalsoplayingamediating role betweenbrand satisfaction and brand loyalty.The strong role of brand affect inthisprocesscanbeattributedtothesampleprofileandtheproductcategoryforwhichtheresearchdatawerecollected.Clothesmaybeperceivedasasocialstatussymbolamonguniversitystudent,whichmayincreasetherelativeeffectoffeelingsondecisions,choiceorjudgmentsaboutthebrand.Futureexaminationoftheresultsabouttheserelationshipshasshownthatbrandtrustandbrandaffectexplain a very satisfactory amount of the observedvariance in brand loyaltywith a 0.70 squaredmultiplecorrelationvalue.Itwasanimportantfindingasoneofthemainpurposesofthestudywastoexaminetheprocessthroughwhichbrandloyaltydevelops.

33

The results also present some important explanations about how consumers would bemotivatedto be ready to pay premiumprices for the brand.Different from some earlier findings (Hermannetal.,2004;Huberetal.,2000,e.g.)indicatingadirectrelationshipbetweensatisfactionandpricetolerance,thepresentresultsshowthatsatisfactionaffectspricetoleranceindirectlythroughbrandaffect,brandtrust,andbrandloyalty.Evaluationofthemodificationindicesstrengthensthisfindingofferingnoadditionallink.AsstatedinH6,consumers’willingnesstopaypremiumpriceforabrandisdirectlyrelatedtohowloyaltheyaretothatbrand.Andthisrelationwasfoundsostrong(estimate6

= 0.73p<0.01)thatbrandloyaltywasabletoexplainmorethanhalfofthevarianceobservedinpricetolerance.Itappearsthattheacceptabilityofpricelevelsforabrandisstronglyinfluencedbyhowstronglyconsumersintendtobuythesamebrandagainandre-patronizeit.

Itisaclearfactthatfirmsthataimnotonlytomaketheirbrandssurviveinintensivelycompetitivemarketsbutalsooutperformtherivalshavetopaygreatattentiontobrandloyalty.Firmscantakeadvantageofpricetoleranceandsomeotherconsequencesiftheysucceedatcreating,maintainingandincreasingbrandloyalconsumerportfolios.Butitisnotalwayseasytoacquireconsumerswillinginalong-termrelationshipwithabrand,especiallyinsomemarketswherenumerousalternativeswithsimilarofferingsexist.

Asfortheantecedents,satisfactionwasseenasthemainfactorintheprocessofestablishingbrandloyalty.But the findings of this research extend some previous insights that satisfactionmay notbeenoughtomakeconsumersloyaltoonebrandoverothers.Inparallelwiththetheoreticalsuggestionsofpreviousstudies,ourcurrentempiricalfindingsalsostress thenecessityofa trustworthyimageforabrandtogainaloyalconsumerportfolio.Anotherimportantfindingistheevidenceabouttherelationshipbetweenconsumption-evokedemotionsandbrandloyalty. Intheirmarketingpoliciescompaniesshouldconcentrateondeepermarketresearchtoacquirecompleteknowledgeaboutthedynamicsofaffectiveresponsestobecapableofprovidingconsumerstofeelsohappy,pleasantandsoforth.Throughtheseeffortscompaniescandevelopconsumerwillingnesstostayinalong-termrelationshipwiththeirbrands.

LIMITATIONSANDFURTHERRESEARCHIMPLICATIONS

Theresultsofthisstudyhaveseverallimitationsthatshouldbeconsidered.Thegeneralizationofthefindingsislimitedbecauseofthelimitedvarietyinthesampleprofile.Anotherlimitationisthatthedatawerecollectedwithrespecttoonlyoneproduct.Intermsoffutureresearch,theselimitationspointtotheneedtoexpandthevarietyofsampleprofilesandproductsonwhichtheresearchisbased.

Thefindingsofthisstudyalsoleadtosomeothersuggestionsforfutureresearch.Theresearchmodelcanberefinedbyevaluatingsomeotherprimaryconsequentsofbrandloyalty.Ontheotherhand,a better understanding of the process inwhich brand loyalty is secured through brand affect canbereintroducedintothemodelasamulti-dimensionalconstructprovidingmoreinformationaboutconsumers’ affective response to brand usage. In addition, future research can provide extendedknowledgeaboutthebrandloyaltyconceptbyalsoconsideringsomecomponentsofthecognitiveprocessinwhichconsumerengageindecision-makingaboutbrands.

34

REFERENCES

Anderson,E.W.(1996).“ConsumerSatisfactionandPriceTolerance,”Marketing Letters,7(3):265-274.

Aaker,D.A.(1996).Building Strong Brand.London:FreePress.

------ (2006). “Measuring Brand Equity across Products and Markets,”California Management Review,38(3):102-120.

Ballester, E.D. andAlleman, J.L.M. (2001). “BrandTrust in theContext ofConsumer Loyalty,”European Journal of Marketing,35(11/12):1238-1258.

------ (2005).“DoesBrandTrustMatter toBrandEquity?”Journal of Brand Management,14(3):187-196.

Bowen, J.T. andChen, S.L. (2001). “TheRelationship betweenCustomerLoyalty andCustomerSatisfaction,”International Journal of Contemporary Hospitality Management,13(5):213-217.

Chaudhuri,A.andHolbrookM.B.(2001).“TheChainEffectsfromBrandTrustandBrandAffecttoBrandPerformance:TheRoleofBrandLoyalty,”Journal of Marketing,65(2):81-93.

DoneyM.P. and Cannon, J.P. (1997). “An Examination of the Nature of Trust in Buyer–SellerRelationships,”Journal of Marketing,61(2):35-51.

Fornell,C. andLarcker,D.F. (1981). “EvaluatingStructuralEquationModelswithUnobservableVariablesandMeasurementError,”Journal of Marketing Research,18(01):39-50.

Garbarino,E.andJohnson,M.S.(1999).“TheDifferentRolesofSatisfaction,TrustandCommitmentinCustomerRelationship,” Journal of Marketing,63(2):70-87.

Gounaris,S.andStathakopoulos,V.(2004).“AntecedentsandConsequencesofBrandLoyalty:AnEmpiricalStudy,”Journal of Brand Management,11(4):283-306.

Hermann, A., Huber, F., Sivakumar, K., andWricke,M. (2004). “An Empirical Analysis of theDeterminantsofPriceTolerance,”Psychology & Marketing,21(7):533.

Homburg,C. andGiering,A. (2001). “PersonalCharacteristics asModeratorsof theRelationshipbetween Customer Satisfaction and Loyalty-An Empirical Analysis,”Psychology and Marketing,18(1):43-66.

Huber,F.,Herrmann,A.,andWricke,M.(2001).“ConsumerSatisfactionAsanAntecedentofPriceAcceptance:ResultsofanEmpiricalStudy,”Journal of Product and Brand Management,10(3):160-169.

35

Jacoby, J. andKyner,D.B. (1973). “BrandLoyalty vs.Repeat PurchasingBehavior,” Journal of Marketing Research,10(01):1-9.

Kalyanaram,G.andLittle,J.D.C.(1994).“AnEmpiricalAnalysisofLatitudeofPriceAcceptanceinConsumerPackageGoods,” Journal of Consumer Research,21(3):408-41.

Lou,G.T.andLee,S.H.(1999).“Consumers’TrustinaBrandandLinktoBrandLoyalty,”Journal of Market Focused Management,4:341-370.

Mittal,B. andLassar,W.M. (1998). “WhyDoCustomersSwitch?TheDynamics ofSatisfactionVersusLoyalty,”The Journal of Services Marketing,12(3):177-194.

Morgan,R.M.andHunt,S.D.(1994).“TheCommitmentTrustTheoryofRelationshipMarketing,”Journal of Marketing,53(3):20-38.

Nyer, P.U. (1997). “A Study of Relationships between Cognitive Appraisals and ConsumptionEmotions,”Academy of Marketing Science Journal,25(4):296-304.

Nunnaly,J.C.(1978).Psychometric Theory.2nded.NewYork:McGrawHill.

OdinY.,Odin,N.,andFlorence,P.V.(2001).“ConceptualandOperationalAspectsofBrandLoyalty:AnEmpiricalInvestigation,”Journal of Business Research,53:75-84.

Oliver,R.L.(1999).“WhenceConsumerLoyalty?”Journal of Marketing,63:33-44.

Quester, P. andLim,A.L. (2003). “Product Involvement /BrandLoyalty: Is There aLink?” The Journal of Product and Brand Management,12(1):22-38.

Ranaweera,C.andPrabhu,J.(2003).“TheInfluenceofSatisfaction,TrustandSwitchingBarriersonCustomerRetentioninaContinuousPurchasingSetting,”International Journal of Service Industry Management,14(4):374-395.

Selnes,F.(1993).“AnExaminationofEffectofProductPerformanceonBrandReputation,SatisfactionandLoyalty,”European Journal of Marketing,27(9):19-35.

------ (1998). “Antecedents and Consequences of Trust and Satisfaction in the Buyer-SellerRelationship,”European Journal of Marketing, 32(3/4):305.

Thiele,S.R.andMackay,M.M.(2001).“AssessingthePerformanceofBrandLoyaltyMeasures,”The Journal of Services Marketing,15(7):529-546.

Westbrook, R.A. (1987). “Product/Consumption-Based Affective Responses and Post PurchaseProcesses,”The Journal of Marketing Research,24(3):258-270.

White,C.andYu,Y.T.(2005).“SatisfactionEmotionsandConsumerBehavioralIntentions,”Journal of Service Marketing,19(6):411-420.