Modeling motivation and habit in driving behavior under lifetime driver's license revocation

(1)

ContentslistsavailableatSciVerseScienceDirect

Accident

Analysis

and

Prevention

jo u r n al h om ep ag e :w w w . e l s e v i e r . c o m / l o c a t e / a a p

Modeling

motivation

and

habit

in

driving

behavior

under

lifetime

driver’s

license

revocation

Chien-Ming

Tseng

a,∗

_,

_Hsin-Li

_Chang

b,1

_,

_T.

_Hugh

_Woo

b,2

a_Ministry_of_{Transportation}_and_{Communications,}₆_Shengfu_Road,_Chung-Shin_Village,_Nantou_City_54045,_Taiwan,_ROC

b_Department_of_{Transportation}_Technology_and_Management,_National_Chiao_Tung_University,₁₀₀₁_University_Road,_Hsinchu_30010,_Taiwan,_ROC

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received8March2011

Receivedinrevisedform5November2012

Accepted20November2012

Keywords:

Licenserevocation

Drivingbehaviors Drivinghabit

Theplannedbehavioraltheory

Theorderedlogitmodel

a

b

s

t

r

a

c

t

Thepurposeofthepresentstudywastoverifythemotivationalfactorsunderlyingthetheoryofplanned behavior(TPB)predictingthedrivingbehavioroflifetimedrivinglicenserevokedoffenders.Ofatotalof

639driverswhoselicenseshadbeenpermanentlyrevoked,544offenderscompletedaquestionnaire

constructedtomeasureattitudestowardbehaviors,subjectivenorms,perceivedbehavioral control,

behavioralintentions(thekeyconstructsoftheTPB),andpreviousdrivinghabitstrength.Theﬁnding

ofthestudyrevealedthatanoffenders’drivingbehaviorafteralifetimelicenserevocationwas

sig-niﬁcantlycorrelatedtobehavioralintention(R=0.60,p<0.01),perceivedbehavioralcontrol(R=0.61, p<0.01),previousdrivinghabit(R=0.44,p<0.01),andattitude(R=0.41,p<0.01).Therewasno evi-dencethatsubjectivenormsincludingroadregulation,societyethics,andpeopleimportanttooffenders

hadaninﬂuenceondrivingbehavior(R=0.03).Lowdrivinghabitstrengthoffendersaremotivatedto

drivebecauseofbehavioralintention,whereasstrongdrivinghabitstrengthoffendersaremotivated

todrivebecauseofperceivedbehavioralcontrol.Previousdrivinghabitstrengthisamoderatorinthe

intention–behaviorrelationship.Themodelappearedsuccessfulwhenprevioushabitswereweak,but

lesssuccessfulwhenprevioushabitswerestrong.

1. Introduction

Manydriversgivenasentenceoflicensesuspension/revocation

(S/R)continuetodrive,butatreducedlevels(Hagenetal.,1980;

Ross and Gonzales, 1988). Ingraham and Waller (1971) found

atleast30%ofdriversgivenS/Rfordrunk-drivingcontinuedto

operateavehicleinspiteofthelicensingaction.Williamsetal.

(1984) concluded that 65% of drivers admitted to operating a

vehicle after a license S/R. Ross and Gonzales (1988) reported

that66%ofthedriverswhoselicensingweresuspendedwerestill

driving.DeYoung(1999) estimatedthat as many as75% ofS/R

licenseddriverscontinuetodrive,althoughtheyapparentlydrove

lessoftenandmorecarefully.Malenfantetal.(2002)showedthe

percentageofmotoristsdrivingwhilesuspendedwas57%ofthe

expectedvalue,relativetotheirrepresentationamongalldrivers.

Changetal.(2006)foundthat86%ofoffenderscontinuedtodrive,

butwithsigniﬁcantlyreduceddrivingactivitiesandmileage.The

generalapproach ofS/R,a driver-based sanction,was intended

∗ Correspondingauthor.Tel.:+886427011026;fax:+886423588039.

E-mailaddresses:tsengcm168@yahoo.com.tw(C.-M.Tseng),

hlchang@cc.nctu.edu.tw(H.-L.Chang),thwoo@mail.nctu.edu.tw(T.H.Woo).

1 _Tel.:₊₈₈₆₃_5731908;_fax:₊₈₈₆₃_5712365.

2 _Tel.:₊₈₈₆₃_5731998;_fax:₊₈₈₆₃_5720844.

tokeep offenders off the road during theirrestrictionperiods.

InthecaseofshorttermS/R,althoughmanyS/Rdriversviolate

drivingrestrictionandcontinuetodrive,mostpriorresearchhas

focusedontheeffectivenessofS/Randsupportedtheviewthat

this is a positive step in reducing subsequentalcohol-involved

drivingbyoffenders(Zadoretal.,1989;HendersonandKedjidjian,

1992; Lund,1993; Sweedlerand Stewart, 1993; NHTSA, 1993).

However,fewstudieshaveexploredmotivationalfactorscausing

offenderstodrivewhile underlicenseS/R.Furthermore,as S/R

is usually awarded for nomore than a few years, few studies

haveexploredS/Rover alongerperiod oftime (Siskind, 1996).

Thus,itseemsthatnostudyhasexploredthemotivationalfactors

underlying the behavior of driving under a long-term S/R. In

thecase of administrative lifetimelicenserevocation (ALLR) in

Taiwan,thebasicgoalistomaintainroadsafetybykeepingsuch

disqualiﬁed drivers off theroad allowing them noopportunity

for rehabilitation in the licensing system design. Chang et al.

(2006)exploredtheeffectivenessofALLR;however,nostudyhas

exploredthemotivationalfactorsofdrivingwhileunderalifetime

license revocation. Therefore, the motivational or suppressive

factorsleadingtodriveunderALLRremainunclear.

Caruseisimportantformanyhouseholdactivitiesinwestern

developedsocietiesaswellasdevelopingcountriessuchasTaiwan.

Householdsusecarsfor travelingtovariousactivities(Eriksson

etal.,2008).Inthelastthreedecades,considerableprogresshas

(2)

beenmadeinexplainingandpredictingtheinitiationofhuman

behaviorsasrevealedbycurrentlypopularattitude-behavior

mod-els(e.g.FishbeinandAjzen,1975;Ajzen,1991).Supportforthe

efﬁcacyofthetheoryofplannedbehavior(TPB)hasbeen

success-fullyfoundinthecontextofcommondrivingbehaviorsrelatingto

roadsafetysuchasseatbeltuse(JonahandDawson,1982;Budd

etal.,1984;ThuenandRise,1994;S¸ims¸eko˘gluandLajunen,2008),

theuseofacarchildrestraintdevice(GodinandKok,1996),

pedes-trianroadcrossingbehavior(EvansandNorman,1998;Moyano

Díaz,2002),speeding(Parkeretal.,1992a;Forward,1997;Elliott etal.,2003; Letirandand Delhomme,2005; DePelsmacker and Janssens,2007;WarnerandÅberg,2008),drunk-driving(Parker etal.,1992a;Åberg,1993;Parkeretal.,1996;Sheehanetal.,1996; GordonandHunt,1998;Marciletal.,2001;Armitageetal.,2002),

aggressivedriving(Parkeretal.,1995,1998;MilesandJohnson,

2003),anddrivingviolations(Parkeretal.,1992b,c,1995;Forward,

2006).TPBhasbeenfoundtobeausefulmodelfororganizingand

understandingpotentialfactorsthatinﬂuenceintentiontoengage

insafedrivingbehaviorandlawcompliance(Yagil,1998;Gordon

andHunt,1998;Poulteretal.,2008).AnextendedTPBseemsa

valu-ableframeworkfor understandingand changingpeople’ssafety

relatedactionsintrafﬁc(Victoiretal.,2005).

AccordingtoTPB, travelmodechoiceis determinedby

atti-tude,subjectivenorm,perceivedbehavioralcontrol,andbehavioral

intention (Ajzen, 1991; Bamberg and Schmidt, 2001, 2003).

Althoughtheseresearcheffortshaveexplainedmoreofthe

reason-based and deliberate natureof behavior, one important aspect

hasbeenoverlooked inresearch,namely,thefactthat manyof

theaforementionedbehaviorsareexecutedonadaily,repetitive

basis,andthereforemaybecomeroutineorhabitual.Lifeconsists

largelyofdailyroutines(HuffandHanson,1986;Pas,1988;Eagly

andChaiken,1993), andtravelmodechoicemayalsobe

deter-minedby habits(Bambergand Schmidt, 2003;Verplanken and

Orbel,2003).Therefore,itisproposedthatwhenbehavioris

per-formedrepeatedlyandbecomeshabitual,thatbehaviorisguidedby

anautomatedcognitiveprocess,ratherthanbeingprecededbyan

elaboratedecisionmakingprocess(i.e.adecisionbasedonattitudes

andintentions)(Aartsetal.,1998).Habithasbeenperceivedasan

automaticlinkbetweenagoalandaspeciﬁcbehavior(Verplanken

andAarts,1999;AartsandDijksterhuis,2000),orasabehavioral

scriptstoredinmemory(SchankandAbelson,1977;Abelson,1981;

Schank,1982;Gärlingetal.,2001;FujiiandGärling,2003).

Habit-ualbehavior,incontrasttomoredeliberatelycontrolledbehavior,

demandsonlya smallamountofattention,andtheindividual’s

control over behavioral intention aswell asthe behavior itself

isminor(Bargh,1994).Habitualcarusehasbeendemonstrated

inseveralstudies(Verplankenetal.,1994,1998;Erikssonetal.,

2008;Heath and Gifford,2002; Klöckneretal., 2003).A strong

habittouseaparticulartravelmodeis,incomparisonwithaweak

habit,characterizedbyseekinglessinformationandaless

elab-oratechoiceoftravelmode(Aartsetal.,1997).Verplankenetal.

(1998)foundthatbothhabitandintentionweresigniﬁcant

predic-torsofcaruseamongdriverswhowereencouragedtodeliberately

thinkabouttravelmodechoice.InaccordancewithTPB,researchers

expectedthatattitudestowardschoosingtouseone’scar,together

withsubjectivenormandperceivedbehavioralcontrol,would

pre-dictbehavioralintentions,whichinturnwereexpectedtopredict

futurebehavior.

Thepresentstudyisconcernedwiththemotivationalfactorsof

drivingbehaviorusingcomponentsofTPBaswellastheprevious

habitualnatureofdriversdrivingactivityunderALLRasan

exam-pleofarepetitivebehavior.ALLRmaybeanexampleofarepetitive

behaviorthatismotivatedbybehavioralintentionandprevious

drivinghabit.Thisstudyhastwogoals.Theﬁrstistoinvestigate

themotivationalfactorsthatcorrelatetoactualdrivingincidence

whileunder ALLR.The secondistoponder therole ofhabitin

attitude-behavior models. For these goals, the researchers

focusedontherelationshipsbetweenbehaviorandreason-based

antecedents(i.e.asdeﬁnedbyTPB)ononehandandbehaviorand

habitstrengthontheotherhand.Thepresentstudyaimedto

exam-inethepsychologicalfactorspredictingtheactualdrivingbehavior

ofoffenderswhohadbeenpunishedbyALLR.Fortheﬁrstgoal,

weadopttheTPBmodel,whichincludeattitude,subjectivenorm,

perceivedbehaviorcontrol,behavioralintention,previousdriving

habitandbehavior.Forthesecondgoal,theresearchersexplored

theTPBmodel’sabilitytopredictbehaviorunderdifferentdriving

habitstrength(i.e.stronghabit,moderatehabit,andweakhabit).

Itisexpectedthatoffenderswithstronghabitwillreporthigher

drivingbehaviorwhileunderALLR.ItisalsoexpectedthattheTPB

modelperformmorepredictabilityforweakhabitthanstronghabit

participants.

2. Method

2.1. Outlineoftheplannedbehaviormodelandprevioushabit

TheTPB,oradaptationsofit,isthemostoftenusedtheoretical

frameworkofmodelsexplainingtrafﬁcsafetybehavior.Inpresent

study,theconceptualmodelisrepresentedinFig.1.TheTPB

con-tendsthatbehavioralintentionstoengageinabehavioristhemain

predictorofactualdrivingbehavior,andthatbehavioralintentions

areinﬂuencedbyattitudestowardsthatbehavior,subjectivenorms

(i.e.,whetherimportant otherswould approveor disapproveof

thebehavior)andperceivedbehavioralcontrol(Ajzen,1991).

Atti-tudesaregenerallydeﬁnedasconsistingofcognitiveandaffective

componentsorantecedents.Inpresentstudy,offendersmay

ratio-nallyunderstandthatdrivingunderALLRisnotgoodorunsafe,

butatthesametimetheymayliketodrive,becauseitgivesthem

agoodfeelingortheyfeeldeservedtodrive.Subjectivenormis

partoftheTPB,andreﬂectstheoffenders’perceivedsocial

pres-sure(whatindividualsbelieveotherpeoplethinktheyshoulddo).

AccordingtotheTPB,theperceivedopinionofsigniﬁcantothers

(subjectivenorm)caninﬂuenceintentionsandbehavior.Perceived

behavioralcontrolrepresentsaneffectonintentiontoperforma

behaviorthatisnotmediatedbyattitudeorsubjectivenorm(Ajzen,

1988;AjzenandMadden,1986).Whilesomeresearchershave

sug-gestedthatcarusemaybehabitual(e.g.Verplankenetal.,1994;

BambergandSchmidt,2003),thepresentstudyhypothesisthat

habitformationleadsto‘automatic’thatmaydeliberatelyleadto

actualdrivingbehavior.Therefore,habitmayactasamoderatorof

theintention–behaviorrelationship.Andintentionsandbehavior

arealsodeterminedbyperceivedbehavioralcontrolandbyhabit

formation.

2.2. Dataandparticipants

DatawerecollectedfromapreviousALLRstudy(Changetal.,

2006)conductedinJuly2003,andafollow-upquestionnaires

sur-veyconductedthreemonthsaftertheﬁrstresponses.Theprevious

ALLR studyinvestigated asample of 768volunteerALLR

offen-derswhohadbeeninvolvedineitherahit-and-runoffencecausing

death/orinjury,oradrunkdrivingoffencecausingdeath/orserious

injury from1993to2002in Taiwan.Only16.8%ofALLR

offen-dersgave up drivingentirely aftertheALLR was imposed.The

samplepopulationofthepresentstudyfocusedonthe83.2%of

ALLRoffenderscontinuingtodrive.SincetheseALLRoffenderswere

expelledfromtheDepartmentofMotorVehicles,objectiverecords

ofdrivingbehaviorsarenotavailable.Thefollow-upquestionnaires

weredirectlymailedtothese639stilldrivingALLRoffendersand

theirself-reportdatawerecollected.Afteratwo-wavetrialed

(3)

AT

SN

PBC

BI B

PDH

Fig.1. Theframeworkoftheplannedbehaviormodelandpreviousdrivinghabit.AT:attitude;SN:subjectivenorm;PBC:perceivedbehavioralcontrol;BI:behavioral

intention;PDH:previousdrivinghabit;B:behavior.

questionnaireswithareturnrateof88%.Finally,atotalof544ALLR

offenderseffectively completedthisfollow-upinvestigationand

usedinthisstudy.Theparticipantsconsistedof98.5%malesand

1.5%females.Over80%werenotcollegeeducated,and70%were

underage40.Ofthisgroupofparticipants,23.9%heldprofessional

driver’slicenses(professionallicenseisdeﬁnedasalicensewhich

qualiﬁesa drivertodedicatedrivingasa job,includingdriving

acar,atruckorabusaccordingtothesedifferentdriving

vehi-cles’requirementsdistinctively)and76.1%heldordinarydriver’s

licenses.

Theself-reportdataonattitudes,subjectivenorms,perceived

behavioral control, behavioral intention, and previous driving

habitswerecollectedfromthepreviousstudy(Changetal.,2006).

Theactualdrivingbehaviorincludingoffenders’drivingfrequency

andannual mileage drivenunder ALLR wasinvestigatedin the

currentfollow-up survey.Allitemswerecodedusing a7-point

scale. Measures of attitudes, subjective norms, and behavioral

intentiontoward driving under ALLR wereconstructed

accord-ingto theprocedures employed byAjzen and Fishbein (1980),

whereasthemeasureofperceivedbehavioralcontrolwasdesigned

accordingtoAjzen’scollaborativework(AjzenandMadden,1986;

Ajzen,1991;BeckandAjzen,1991).Themeasureofactualdriving

behaviorwasself-reportedbytheparticipantsconcerningdriving

frequency.Self-reportdrivingfrequencywasdeﬁnedfrom

very-highfrequencytovery-lowfrequency.

Table1outlinestheobservedvariablesandquestionitemsofthe

presentstudy.Themeasureofattitudeswerebasedonfour

seman-ticdifferentials(e.g.,goodor bad;safeorunsafe),and assessed

therespondent’spositiveornegativeevaluationofdrivingunder

ALLR.Thefouritemswereaveragedtoobtainthescalescore.The

internalconsistencyofthescalewas0.88.Thesubjectivenorms

weremeasuredbysixquestionsassessingtherespondent’s

eval-uationofhisreferentopinionsregardingactualdrivingbehavior

in the described situation. The averageof the six items

corre-spondstothescoreofthisscale.Cronbach’salphawas0.82.The

perceivedbehavioral controlconstructincludedfouritems that

wereaveragedtoobtainthescalescore. Thismeasureassessed

therespondent’sperceptionofcontroloverdrivingafterALLR.The

internalconsistencyofthisscaleaccordingtoCronbach’salphawas

0.93.Themeasureofbehavioralintentionincludedtwoitems

mea-suringthestrengthoftherespondent’sintentiontodriveafterALLR.

Responsestothetwoitemswereaveragedtoobtainthescoreon

thescale.InternalconsistencyasmeasuredbyCronbach’salpha

was0.92. Theresearchersoperationalizedthemeasurement for

habitstrengthsothat themeasurement would matchthe

con-ceptofageneralizedtypeofhabit.Thus,theresponse-frequency

(RF)measureofgeneralhabitstrengthwhichwasusedand

vali-datedinanumberofearlierstudies(Aarts,1996;Aartsetal.,1997;

Verplankenetal.,1998)wasadopted.WhileVerplankenandAarts

(1999)suggestedthatanimportantfeatureofhabitualbehavior

isautomaticityofresponding,presentparticipantsrespondedto

10itemsthatvariedwidelyintravelpurposes(e.g.goingtothe

market,visitingfriends/orrelatives).Theresearchersassumedthat

themoreinvariantparticipants’responseswerethestrongerthe

habitstrengthwasthusindexedbythemeanofthe10items.The

measureof habitstrength hadhighinternal consistency

(Cron-bach’salpha=0.94)whichrevealedthereliabilityisgood.Inorder

tojustifythe variationis a good representationof thehabitual

behavior.Theresearchersfurthertoconﬁrmweatherthevalidity

isacceptableornot.Afterconductingthe“Principalcomponents

analyses”,atwo-dimensionalstructureappeared.Thereweretwo

eigenvaluesgreaterthan1(6.601and1.070).Theﬁrstcomponent

includinghabitmeasureitemsfrom1to8accountedfor55.803%

ofthevarianceandthesecondcomponentincludinghabit

mea-sureitemsfrom9to10accountedfor 20.907%ofthevariance,

totallyaccountedfor76.710%ofthevariance.Finally,itisconcluded

thatthisvariationinpresentstudyisagoodrepresentationofthe

habitualbehavior.

3. Results

3.1. CorrelationsofmotivationaldrivingfactorsunderALLR–

combinetheplannedbehaviortheoryandhabit

Correlationswerecalculated betweendriving attitudes,

sub-jectivenorms,perceivedbehavioralcontrol,behavioralintention,

previousdrivinghabit,andactualdrivingbehaviormeasures.These

correlationsareshowninTable2.Theresultsindicatethatactual

drivingbehaviorwascorrelatedwithattitude(R=0.41, p<0.01),

perceivedbehavioralcontrol(R=0.61,p<0.01),previousdriving

habit(R=0.44,p<0.01),andbehavioralintention(R=0.60,p<0.01).

Amongthesesigniﬁcantcorrelativevariables,perceivedbehavioral

controlandbehavioralintentionwerethetwomostimportant

fac-torscorrelatedwithactualdrivingbehavior.Theresultsindicate

thattherespondentsthinkdrivingacarisaneed,andwhenever

needingtodrive,respondentswilldriveevenifalicensehadbeen

revoked.Also,mostrespondentsthinkdrivingunderALLRiseasy

andpossible.Attitudeandpreviousdrivinghabitwerealso

signiﬁ-cantlycorrelatedwithactualdrivingbehavior.Theresultsindicated

thatbehavioralintentionwassigniﬁcantlycorrelatedwithattitude

(R=0.36,p<0.01),perceivedbehavioralcontrol(R=0.65,p<0.01),

andpreviousdrivinghabit(R=0.47,p<0.01)suchthatthe

respon-dentswhoreportedexperiencingamorepositiveaffect,moreneed

todrive,andhigherpriordrivinghabitstrengthoverdrivingunder

ALLR,alsoreportedstrongerintentions.However,theresults

indi-catethatsubjectivenormsarenotrelatedtobehavioralintention

(R=0.07)aswellasactualdrivingbehavior(R=0.03).Nomatter

(4)

Table1

Observedvariablesandquestionitems(N=544).

Factors Item Cronbach’s

alpha

Questions Scoring

Behavior 1 – Althoughyourlicensewasrevoked,youdrovewith frequencyafterALLR. 7=veryhigh,1=verylow Behavioralintention 2 0.92 Althoughyourlicensewasrevoked,you havetodriveinthefuture. 7=extremelylikely,

1=extremelyunlikely Althoughyoulicensewasrevoked,you willdriveinthefuture.

Attitude 4 0.88 YouthinktodriveunderALLRis good. 7=extremelylikely,

1=extremelyunlikely YouthinktodriveunderALLRis convenient.

YouthinktodriveunderALLRis deserved. YouthinktodriveunderALLRis safe.

Subjectivenorm 6 0.82 Mostpeopleimportanttoyouwould agreeyoutodrive. 7=extremelylikely, 1=extremelyunlikely Thosewhoimportanttoyou,theiropinionhave theinﬂuenceonyou.

Yoursocieties agreeyoutodriveunderALLR.

Yoursocietieshave inﬂuenceonyoufordrivingunderALLR. Regulations agreeyoutodrive.

Regulations inﬂuenceyoutodrive.

Perceivedbehavioralcontrol 3 0.93 DrivingunderALLRis easyforyou. 7=extremelylikely, 1=extremelyunlikely DrivingunderALLRis possibleforyou.

Wheneveryouhavetogosomewhere,youwill choosedrivingacar.

Previousdrivinghabita ₁₀ _0.94 _You _drive_a_car_to_go_around_somewhere_market. ₇₌_extremely_likely,

1=extremelyunlikely You driveacartovisityourfriends/orrelatives.

You driveacarforshopping.

You driveacartohaveyourdinner.

You driveacartotransportyourkidsorfamilies.

You driveacartogoaroundsomewheredowntown.

You driveacarforleisureactivitiesonweekdays. You driveacarforleisureactivitiesonweekends.

You driveacarforcommutingtowork.

You driveacarforbusiness.

a_Participants_were_requested_to_response_as_soon_as_possible,_and_not_think_too_much_or_take_too_much_time_to_respond.

theseperceptionshadnoinﬂuenceonbothbehavioralintention

andactualdrivingbehavior.

3.2. MotivationaldrivingfactorsunderALLR

According tothe planned behavior theory, driving behavior

under ALLR should be predicted by motivational components.

Giventhatthedependentvariable“drivingfrequencyunderALLR”

hasanordinalnature,withsevenpossibleoutcomes(from1=very

lowto7=veryhigh).Ordinaryleastsquaresareseldomappropriate

forsuchdatasinceitrequiresacontinuousdependentvariable.An

ordered-responsemodelappearsthemostappropriateapproach.

Theordered-discretechoicemodels(i.e.,theorderedlogitmodes)

havebeenemployedbyseveralresearchers(see,forexample,Pai

and Saleh, 2008; Neyens and Boyle,2008)for modeling injury

severities.Therefore,theorderedlogitmodelwasadoptedto

esti-matethemotivationalfactorsofdrivingbehaviorunderALLR.The

theoreticalframeworkoftheorderedlogitmodelandmethodof

evaluationhasthoroughlydiscussedinseveralstudies(e.g.,Long,

1997; Borooah, 2001). The actual driving behavior under ALLR

was modeled as a function of the attitudes, subjective norms,

perceivedbehavioral control, behavioralintention and previous

driving habits, wascreated and veriﬁed the motivational

driv-ingfactorsunderALLR.Theresearchersexaminedwhetheractual

drivingbehaviorformedasa resultofthevariables ofTPB.The

results presented in Table 3 indicate that attitudes, perceived

behaviorcontrol,behavioralintentionandpreviousdrivinghabits

accountforaround49%(PseudoR2₌_0.487)_of_the_total_variance_in

actualdrivingbehaviors.Acloserlookattheseresultsindicatedthat

behavioralintention(ˇ=0.718,p=0.000)andperceivedbehavioral

control(ˇ=0.651,p=0.000)arethetwomostimportantfactors

inthedeterminationofactualdrivingbehavior.Previousdriving

habits(ˇ=0.141,p=0.049)andattitudes(ˇ=0.266,p=0.004)also

signiﬁcantlyinﬂuenceactualdrivingbehavior.Theresultsindicate

thatsubjectivenorms(ˇ=−0.022,p=0.841)arenotsigniﬁcantly

associatedwithactualdrivingbehavior.

3.3. MotivationaldrivingfactorsunderALLR– TPBmodelwith

differentdrivinghabitstrengths

Inthissection,theeffectsoftheTPBmodelwhileunder

vari-ousdrivinghabitstrengthsareexplored.Previousstudiesclassiﬁed

participant’shabitstrengthintodifferentlevelsmainlybyuseof

theparticipants’standarddeviation(i.e.atonestandarddeviation

belowthemean:weakornohabit;atthemean:moderatehabit;

andatonestandarddeviationabovethemean:stronghabit;Aiken

andWest,1991;Gardner,2009).Inthepresentstudy,the

respon-dent’shabitstrengthwasdividedbysequenceintothreegroups

Table2

Correlationamongtheoryofplannedbehaviorvariables(N=544).

Studyvariable Mean S.D. 1 2 3 4 5 6

1.Behavior 4.7 1.9 –

2.Behavioralintention 6.1 1.2 0.60** _–

3.Attitudes 3.8 1.0 0.41** _0.36** _–

4.Subjectivenorms 3.9 0.7 0.03 0.07 −0.04 –

5.Perceivedbehavioralcontrol 5.7 1.4 0.61** _0.65** _0.43** _0.002 _–

6.Previousdrivinghabits 4.8 1.4 0.44** _0.47** _0.49** _−0.023 _0.47** _–

(5)

Table3

Orderedlogitmodelestimationresults:predictionofbehaviorbyintension,attitudes,subjectivenorms,PBCandpreviousdrivinghabits–allparticipants,N=544,average

annualmileage=14,370km.

Variables ˇ S.E. Wald Signiﬁcance(P) 95%C.I.

Intension 0.718*** _0.098 _53.782 _0.000 _0.526_to_0.910

Attitudes 0.266*** _0.093 _8.126 _0.004 _0.083_to_0.448

Subjectivenorms −0.022 0.110 0.040 0.841 −0.237to0.193

PBC 0.651*** _0.084 _60.730 _0.000 _0.487_to_0.815

Previousdrivinghabits 0.141* _0.072 _3.851 _0.049 _0.000_to_0.282

NagelkerkePseudoR2₌_0.487;_Cox_&_Snell_Pseudo_R2₌_0.475;_{Log-likelihood:}_1674.296;_ˇ:_regression_{coefﬁcients;}_S.E.:_standard_error_of_ˇ. * _p_<_0.05.

**_p_<_0.01. ***_p_<_0.001.

Table4

Orderedlogitmodelestimationresults: prediction ofbehavior byintension, attitudes,subjectivenorms andPBC– highhabitstrength, n=180,averageannual

mileage=16,945km.

Intension 0.439* _0.205 _4.571 _0.033 _0.037_to_0.841

Attitudes 0.136 0.150 0.820 0.365 −0.158to0.431

PBC 0.563*** _0.145 _15.067 _0.000 _0.279_to_0.847

NagelkerkePseudoR2₌_0.241;_Cox_&_Snell_Pseudo_R2₌_0.231;_{Log-likelihood:}_439.195;_ˇ:_regression_{coefﬁcients;}_S.E.:_standard_error_of_ˇ. * _p_<_0.05.

**_p_<_0.01. ***_p_<_0.001.

Table5

Orderedlogitmodelestimationresults:predictionofbehaviorbyintension,attitudes,subjectivenormsandPBC–moderatehabitstrength,n=184,averageannual

mileage=13,258km.

Intension 0.782*** _0.168 _21.713 _0.000 _0.453_to_1.112

Attitudes 0.579** _0.184 _9.917 _0.002 _0.219_to_0.940

PBC 1.046*** _0.177 _35.076 _0.000 _0.700_to_1.392

NagelkerkePseudoR2₌_0.420;_Cox_&_Snell_Pseudo_R2₌_0.407;_{Log-likelihood:}_614.850;_ˇ:_regression_{coefﬁcients;}_S.E.:_standard_error_of_ˇ. *_p_<_0.05.

** _p_<_0.01. ***_p_<_0.001.

witheachgroupcontainingasimilarnumberofrespondents.The

reasonfor this isthat thegroupingwillreduce thevariance of

therespondents’habitstrength.Iftheseresultsrepresenthabit

strengthinﬂuenceonbehavior,thentherelationshipofhabitand

behavior can be more easily conceived. Therefore, all

respon-dentswereclassiﬁedintothreehabitstrengthgroupsincluding

a high driving habit strength group, a moderate driving habit

strengthgroup,andaweakdrivinghabitstrengthgroup.Tables4–6

revealtheactualdrivingbehaviorsregressedonattitude,perceived

behavioralcontrol,andbehavioralintentionforthethreedriving

habitstrengthgroups.Forthehighhabitstrengthgroup,perceived

behavioralcontrol(ˇ=0.563,p=0.000)isthemostimportant

fac-torthatinﬂuencedactualdrivingbehavior.However,forthelow

habitstrengthgroup,behavioralintention(ˇ=0.822,p=0.000)is

themostdeterminantfactoronactualdrivingbehavior.Attitudes

aresigniﬁcantly associatedwiththeparticipants’actualdriving

behaviorformoderatehabitstrength(ˇ=0.579,p=0.002)andlow

habitstrength(ˇ=0.401,p=0.009)groups.Therewasnoevidence

thatsubjectivenormshaveaninﬂuenceontherespondents’

driv-ingbehaviornomatterwhetherdrivinghabitsarestrongornot

(ˇ=−0.038,p=0.819;ˇ=−0.087,p=0.631;ˇ=212,p=0.412for

strong,moderateandlowhabitstrength,respectively).

The strength of the associations with actual driving

behav-ior was correlated with perceived behavior control, behavioral

intention and attitude, but not the norms in all three habit

strengthgroups.Theassociationbetweenbehavioralintentionand

actualdrivingbehaviorshouldbecomeweakerwhiledrivinghabit

strengthwashigh.Theassociationbetweenbehavioralintention

Table6

Orderedlogitmodelestimation results: prediction ofbehavior by intension,attitudes, subjectivenorms and PBC– low habitstrength,n=180,averageannual

mileage=4,349km.

Intension 0.822*** _0.153 _29.013 _0.000 _0.523_to_1.121

Attitudes 0.401** _0.154 _6.786 _0.009 _0.099_to_0.703

Subjectivenorms 0.212 0.258 0.674 0.412 −0.294to0.718

PBC 0.469*** _0.128 _13.466 _0.000 _0.218_to_0.719

NagelkerkePseudoR2₌_0.593;_Cox_&_Snell_Pseudo_R2₌_0.582;_{Log-likelihood:}_566.727;_ˇ:_regression_{coefﬁcients;}_S.E.:_standard_error_of_ˇ. *_p_<_0.05.

** _p_<_0.01. ***_p_<_0.001.

(6)

andactualdrivingbehaviorbecomestrongerwhilehabitstrength

wasweak.WhiletakingacloserlookatthePseudoR2_of_the_three

models,0.241,0.420and0.593forhigh,moderateandlowhabit

strengthgroup,respectively,theTPBmodelaccountsforahigher

percentageofthetotalvarianceoftheactualdrivingbehaviorwhile

previoushabitstrengthwasweak,andviceversa.

4. Discussion

Whilefacingaserioustrafﬁcviolationproblem,trafﬁc

authori-tiescommonlyadoptadeterrentofincreasingthepenaltyforsuch

serioustrafﬁcoffence.Thepenaltiesincludecar-basedsanctions

and driver-basedsanctions.In driver-basedsanctions,themost

seriouspenaltyforanoffenderisalifetimedriver’slicense

revo-cation.Suchadeterrenthastheaimofmaintainingroadsafetyby

suspendingserioustrafﬁcoffenders,andofferingnoopportunity

fortheoffendertore-enterthelicensingsystemevenwhenthey

demonstrateanabilityandwillingnesstofollowtheregulations

ofroadandsociety.However,inamodernsocietymanyeconomic

andsocialactivitiesrelyondrivingavehicletofulﬁllliving

pur-poses.Manypreviousstudies,basedonashort-termlicenseS/R,

havefoundthatdriverscontinuedtodrivebutatreducedlevels

(Hagenetal.,1980;RossandGonzales,1988;SmithandMaisey, 1990;Voasand DeYoung,2002).In thecaseoflifetimedriver’s

licenserevocation,driversareinthesamesituationasashort-term

licenseS/R,andhavetoseekanalternativeefﬁcientwayofgoing

towork,shopping,visitingrelatives/friends,andforleisure

pur-poses.Suchanalternativesystemoftransportationmustcontinue

forarelativelylongertime,insomecases,therestoflife.Toendure

suchalongalternativesystemoftransportationismoredifﬁcult

thanthatofashortersuspensionperiod.Therefore,thehypothesis

thattheratiooflifetimelicenserevocationoffenderscontinuingto

driveisgreaterthantheratioforashort-termlicenserevocationis

alogicalassumption.Changetal.(2006)foundthatlifetimedriver’s

licenserevocationoffenderswhocontinuedtodrive(83.2%)were

higherthanthatinpreviousﬁndingsallowingtheconclusionthat

complyingwithashort-termlicenseS/Rmayberelativelyeasy

formostpeoplewhilealifetimesuspensionofdrivingprivileges

maybetoomuchfordriverstoendure.However,thereisalmost

nostudyexploredthemotivationalfactorinﬂuenceonoffender’s

driving,nomatterunderashort-termS/R,along-termlicenseS/R

orlifetimelicenserevocation.PresentstudyappliedTPBmodelas

wellashabitstrengthtospeciﬁcinstancesofaberrantbehaviorand

successfullyexploredthemotivationalfactors,whichinﬂuence

off-enderstodriveavehicleaftertheirlicenseshadbeenrevokedfor

lifelong.

Thepresent studyresultsrevealedthatthesigniﬁcant

moti-vational factors correlated to driving afterALLR are behavioral

intention(R=0.61), perceivedbehavioral control (R=0.60),

pre-viousdrivinghabit(R=0.44),andattitude(R=0.41); theformer

two factorshavingthehighest andthe secondhighest

inﬂuen-tialroles.ThisﬁndingisconsistentwithwhattheTPBcontends

inthatintentions toengageinabehavioristhemainpredictor

of actualbehavior, and behavioral intentions are inﬂuenced by

attitudestowardsthatbehavior, subjectivenorms(i.e.,whether

importantotherswouldapproveordisapproveofthebehavior),

and perceived behavioral control (Ajzen, 1991). Empiricaltests

ofintention–behavior correlativestudiesshowthat intentionis

reliablyassociatedwithbehavior(i.e.,WebbandSheeran,2006).

Armitageand Conner (2001),in a meta-analysisof 185studies

thathaveusedtheTPB,alsofoundthatthesample-weighted

aver-agecorrelationbetweenmeasuresofintentionandbehaviorwas

0.47.Theintention–behaviorcorrelationofthepresentALLRcaseis

higherthantheaverageofpreviousﬁndings.Moreover,thepresent

studyisbasedonoffendersofserioustrafﬁcoffences.

Perceivedbehavioralcontrolisthesecondstrongestinﬂuential

factormotivatingactualdrivingbehaviorafterALLR.Presentresults

indicatedthatrespondentsdeemdrivingafterALLRisaneed,and

whendriversperceiveaneedtodriveordesiretoaccomplish

liv-ingactivitiesviadrivingavehicle,respondentswilldrive.Attitudes

werealsoasigniﬁcantfactorinﬂuencingactualdrivingbehavior.

ManyrespondentsthinkdrivingunderALLRissafeandconvenient

andcontinuetodrive.Attitudeusuallyplaysanimportantrolein

caruseandtravelmodechoiceastheﬁndingsofthisstudyindicate

andareconsistentwithpreviousﬁndings(Ajzen,1991;Tertoolen

etal.,1998).Thepresentstudyisalsoconsistentwiththeﬁndings

ofpreviousstudiesinthattherateoftrafﬁclawviolationsis

inﬂu-encedbytheleveloflawenforcement,andaslongastheperceived

threatofdetection remainslow, trafﬁcviolatorsareunlikely to

changebehavior(Harper,1991;Yagil,1998).Subjectivenormsare

notstronglycorrelatedwithactualdrivingbehavior(R=0.03)and

behavioralintention(R=0.07).Roadregulations,socialethics,and

importantpeopletooffendershavealmostnoinﬂuenceon

choos-ingtooperateavehicle.Thisﬁndingisdifferentfromotherprevious

trafﬁcsafetystudies,e.g.LajunenandRäsänen(2004),whofound

thatthesubjectivenormwasthestrongestpredictorofthe

inten-tionto usea bicycle helmet.This ﬁndingisalsodifferent from

Manstead’s(2000)reviewof20studiesﬁndingonlythree

repor-tingnoempiricalsupportfortheindependentpredictiveutility

ofpersonalnorm.Thepresentmodelbetterpredictstheseresults

thanpreviousstudies,whichpredictbetween23%and47%ofthe

variance(Parkeretal.,1992a,1995).

Speciﬁcally,for thelow drivinghabitgroup,themost

moti-vational factor inﬂuencing drivers to drive after an ALLR was

behavioralintention,whereasforthestrongdrivinghabitgroup,

the most inﬂuential factor inﬂuencing drivers to drive was

perceivedbehavioralcontrol.Behavioralintentioninﬂuencedthe

lowdrivinghabitgroupmorethanthehighdrivinghabitgroup.

Incontrast,perceivedbehavioralcontrolinﬂuencedthehighand

moderatedrivinghabitgroupsmorethan thelow drivinghabit

group.TheresultsofthisstudyareconsistentwiththeTPB

empha-sisofthereason-basedantecedentsofbehavior.However,when

behavior is repeatedlyand satisfactorily executedand becomes

habitual,itmayloseitsreasonedcharacter.Behaviormaythenbe

moreguidedbytheautomaticityofstimulus–responserelations

andlessbyintentions(Verplankenetal.,1998),therefore,

inten-tionsareassumedtopredictbehaviortotheextentthatthehabit

componentisweak,andnot,ortoalesserdegree,whenhabitis

strong(Triandis,1977).

AccordingtoWoodandcolleagues(OuelletteandWood,1998;

Woodetal.,2002;WoodandQuinn,2005),behaviorsthatare

per-formedfrequentlyinstablecontextssupportthedevelopmentof

habits,and thus theimpactofintention onbehavioris

attenu-ated.Ameta-analysisbyOuelletteandWood(1998)showedthat

when behavior is practiced repeatedly and thecontext of

per-formance isstable,past behavioris a betterpredictor offuture

behavior than is intentionwhereas the reversewas truewhen

behaviorswereperformedinfrequentlyinunstablecontexts.

Sim-ilarly,Verplankenetal.(1998)foundaninteractionbetweenhabit

andintentionsuchthatintentionswereonlysigniﬁcantlyrelated

tobehaviorwhenhabitstrengthwasweak.Whenparticipants

pos-sessedstronghabits,intentionshadlessinﬂuenceonsubsequent

behavior(seealsoFergusonandBibby,2002;Klöckneretal.,2003);

seeAjzen(2002)foradifferentview.Thus,whetherbehaviorshave

thepotentialtobecontrolledbyhabitcouldbeanimportant

mod-eratorofintention–behaviorrelations.Thepresentstudyalsofound

drivinghabitstrengthwasamoderatorintheintention–behavior

relation,moreover, themodel appeared relatively successfulin

modelingdrivingbehaviorunderALLRwhenprevioushabitswere

weak,whereas lesssuccessfulinmodelingwhenprevioushabit

(7)

5. Conclusions

PreviousresearchfoundthattheALLRpolicyeffectively

inﬂu-enced offenders to drive less frequently, fewer miles, more

carefully,moredefensively(Changetal.,2006)andlesscrashrisk

(Changetal.,2011).Thepresentstudyidentiﬁedthemotivational

factorsleadingoffenderstodrivewitharevokedlicensed.However,

theprivilegetodriveisvalued, andwithdrawaloftheprivilege

feared(DeYoungand Gebers,2004).Evenfor a short-termS/R,

one-ﬁfthoftheUSstatesrejectedtheadoptionofadministrative

S/Rbecauseitcouldleadtolossofemployment,inturn

impact-ingtheoffender’sdependentsandsubsequentsocialwelfarecosts

(KnoebelandRoss,1997;VoasandDeYoung,2002).ALLRmaynot

beimplementedindevelopedcountries;however,itmaybe

imple-mentedindevelopingcountries.FuturestudyemploysTPBmodel

orothermeasuringmethodsmayconﬁrmpresentresults.In2001,

theTaiwanConstitutionalCourtaskedthetransportationauthority

toreconsiderwhetherALLRoffendersshouldbeallowedtore-enter

thelicensingsystemiftheycandemonstrateanabilityand

willing-nesstofollowtheregulationsoftheroadandsociety.In2006,the

ALLRpolicywasrevisedbyalicenserevocationfor8,10,or 12

yearsaccordingtothesameoffencecausinginjury,seriousinjury,

ordeath,respectively.

Acknowledgements

Thisstudy was approved by the Ministryof Transportation

andCommunication,Taiwanandconductedwiththeassistanceof

DepartmentofMotorVehicles;participantswerevolunteersand

withoutpayment; ﬁnancewassupportedbytheDepartment of

TransportationTechnologyandManagement,NationalChiaoTung

University.

References

Aarts,H.,1996.Habitanddecisionmaking:thecaseoftravelmodechoice. Unpub-lishedDoctoralDissertation,UniversityofNijmegen,TheNetherlands. Aarts,H.,Verplanken,B.,VanKnippenberg,A.,1997.Habitandinformationusein

travelmodechoices.ActaPsychologica96,1–14.

Aarts,H.,Verplanken,B.,VanKnippenberg,A.,1998.Predictingbehaviorfrom

actionsinthepast:repeateddecisionmakingoramatterofhabit?Journalof

AppliedSocialPsychology28(15),1355–1374.

Aarts,H.,Dijksterhuis,A.,2000.Habitsasknowledgestructures:automaticityin goal-directedbehavior.JournalofPersonalityandSocialPsychology78,53–63. Abelson,R.P.,1981.Psychologicalstatusofthescriptconcept.AmericanPsychologist

36,715–729.

Åberg,L.,1993.Drinkinganddriving:intentions,attitudes,andsocialnormsof

Swedishmaledrivers.AccidentAnalysisandPrevention25,289–296.

Aiken,L.S.,West,S.G.,1991.Multipleregression:testingandinterpreting interac-tions.SagePublications,NewburyPark.

Ajzen,I.,1988.Attitudes,personalityandbehavior.OpenUniversityPress,Milton

Keynes,UK.

Ajzen,I.,1991.Thetheoryofplannedbehaviour.OrganizationalBehaviourand

HumanDecisionProcesses50,179–211.

Ajzen,I.,2002.Residualeffectsofpastonlaterbehavior:habituationandreasoned actionperspectives.PersonalityandSocialPsychologyReview6,107–122. Ajzen,I.,Fishbein,M.,1980.UnderstandingAttitudesandPredictingSocialBehavior.

Prentice-Hall,NewJersey.

Ajzen,I., Madden,T.J., 1986. Predictionof goal-directedbehaviour: attitudes,

intentions,andperceivedbehaviouralcontrol.JournalofExperimentalSocial

Psychology22,453–474.

Armitage,C.J.,Conner,M.,2001.Efﬁcacyofthetheoryofplannedbehaviour:a meta-analyticreview.BritishJournalofSocialPsychology40,471–499.

Armitage,C.J.,Norman,P.,Conner,M.,2002.Canthetheoryofplannedbehaviour mediatetheeffectsofage,genderandmultidimensionalhealthlocusofcontrol? BritishJournalofHealthPsychology7,299–316.

Bamberg,S.,Schmidt,P.,2001.Theory-drivensubgroup-speciﬁcevaluationofan

interventiontoreduceprivatecaruse.JournalofAppliedSocialPsychology31, 1300–1329.

Bamberg,S.,Schmidt,P.,2003.Incentives,moralityorhabit?Predictingstudents’ caruseforuniversityrouteswiththemodelsofAjzen,SchwartzandTriandis.

EnvironmentandBehavior35,264–285.

Bargh,J.A.,1994.Environmentalcontrolofgoal-directedaction:automaticand

strategiccontingenciesbetweensituationsandbehavior.In:Nebraska

Sympo-siumonMotivation,vol.41.NewYorkUniversity,USA,pp.71–124.

Beck,L.,Ajzen,I.,1991.Predictingdishonestactionsusingthetheoryofplanned behaviour.JournalofResearchinPersonality25,285–301.

Borooah,V.,2001.Logitandprobit:orderedandmultinomialmodels.In:Sage Uni-versityPapersSeriesonQuantitativeApplicationintheSocialSciences07-138.

SagePublications,ThousandOaks,CA.

Budd,R.J.,North,D.,Spencer,C.,1984.Understandingseat-beltuse:atestofBentler andSpeckart’sextensionofthetheoryofreasonedaction.EuropeanJournalof SocialPsychology14,69–78.

Chang,H.L.,Woo,T.H.,Tseng,C.M.,2006.Isrigorouspunishmenteffective?Acase studyoflifetimelicenserevocationinTaiwan.AccidentAnalysisandPrevention

38,269–276.

Chang,H.L.,Woo,T.H.,Tseng,C.M.,Tseng,I.Y.,2011.Drivingbehaviorsand acci-dentriskunderlifetimelicenserevocation.AccidentAnalysisandPrevention

43,1385–1391.

DePelsmacker,P.,Janssens,W.,2007.Theeffectsofnormsattitudesandhabits

onspeedingbehavior:scaledevelopmentandmodelbuildingandestimation.

AccidentAnalysisandPrevention39,6–15.

DeYoung,D.J.,1999. Anevaluationofthespeciﬁc deterrenteffectsof vehicle

impoundmentonsuspended,revoked,andunlicenseddriversinCalifornia. Acci-dentAnalysisandPrevention31,45–53.

DeYoung,D.J.,Gebers,M.A.,2004.Anexaminationofthecharacteristicsand traf-ﬁcrisksofdriverssuspended/revokedfordifferentreasons.JournalofSafety

Research35,287–295.

Eagly,A.H.,Chaiken,S.,1993.ThePsycologyofAttitudes.HarcourtBraceJovanovich, FortWorth,TX.

Elliott,M.A.,Armitage,C.J.,Baughan,C.J.,2003.Drivers’compliancewithspeed

limits:anapplicationofthetheoryofplannedbehavior.JournalofApplied

Psychology88,964–972.

Eriksson,L.,Garvill,J.,Nordlund,A.M.,2008.Interruptinghabitualcaruse:the impor-tanceofcarhabitstrengthandmoralmotivationforpersonalcarusereduction. TransportationResearchPartF11,10–23.

Evans,D.,Norman,P.,1998.Understandingpedestrians’roadcrossingdecisions:an applicationofthetheoryofplannedbehaviour.HealthEducationResearch13, 481–489.

Ferguson,E.,Bibby,P.A.,2002.Predictingfutureblooddonorreturns:pastbehavior, intentions,andobservereffects.HealthPsychology21,513–518.

Fishbein,M.,Ajzen,I.,1975.Belief,Attitude,Intention,andBehavior:AnIntroduction

toTheoryandResearch.Addison-Wesley,Reading.

Forward,S.E.,1997.Measuringattitudesandbehaviourusingthetheoryofplanned behaviour.In:Rothengattter,T.,Vaya,E.C.(Eds.),Trafﬁc&TransportPsychology.

Pergamon,NY,pp.353–365.

Forward,S.E.,2006.Theintentiontocommitdrivingviolations-Aqualitativestudy. TransportationResearchPartF9,412–426.

Fujii,S.,Gärling,T.,2003.Developmentofscript-basedtravelmodechoiceafter forcedchange.TransportationResearchPartF6,117–124.

Gardner,B.,2009.Modellingmotivationandhabitinstabletravelmodecontexts. TransportationResearchPartF12,68–76.

Gärling,T.,Fijii,S.,Boe,O.,2001.Empiricaltestsofamodelofdeterminantsof script-baseddrivingchoice.TransportationResearchPartF4,89–102.

Godin,G.,Kok,G.,1996.Thetheoryofplannedbehavior:areviewofits

applica-tionstohealth-relatedbehaviors.AmericanJournalofHealthPromotion11,

87–98.

Gordon,C.,Hunt,M.,1998.Thetheoryofplannedbehaviorappliedtospeeding,

drink-drivingandseatbeltwearing.In:RoadSafetyConferenceProceedings. Hagen,R.E.,McConnell,E.J.,Williams,R.L.,1980.Suspensionandrevocationeffects

ontheDUIoffender.CaliforniaDepartmentofMotorVehicles,Sacramento,CA. Harper,J.G.,1991.Trafﬁcviolationdetectionanddeterrence:implicationsfor

auto-maticpolicing.AppliedErgonomics22,189–197.

Heath,Y.,Gifford,R.,2002.Extendingthetheoryofplannedbehavior:

predic-tiontheuseofpublictransportation.JournalofAppliedSocialPsychology32, 2154–2189.

Henderson,A.,Kedjidjian,C.,1992.Administrativelicenserevocation–arewe driv-ingdrunksofftheroad?TrafﬁcSafety,6–8.

Huff,J.O.,Hanson,S.,1986.Repetitionandvariabilityinurbantravel.Geographical Analysis22,70–93.

Ingraham,W.S.,Waller,J.A.,1971.Alcohol-impaireddriving,licensesuspensions,

andtransportationneedsduringintoxicationorsuspensionamongalcoholics.

CrashReportNo.IV-1,DepartmentofMentalHealth,Waterbury.

Jonah,B.A.,Dawson,N.E.,1982.Predictingseatbeltusefromattitudinaland

nor-mativefactors.AccidentAnalysisandPrevention14,305–310.

Klöckner,C.A.,Matthies,E.,Hunecke,M.,2003.Problemsofoperationalizinghabits andintegratinghabitsinnormativedecision-makingmodels.JournalofApplied

SocialPsychology33,396–417.

Knoebel,K.Y.,Ross,H.L.,1997. Effectsof administrativelicenserevocationon

employment.AccidentAnalysisandPrevention29(5),595–611.

Lajunen,T.,Räsänen,M.,2004.Cansocialpsychologicalmodelsbeusedtopromote

bicyclehelmetuseamongteenagers?AcomparisonoftheHealthBeliefModel,

TheoryofPlannedBehaviorandtheLocusofControl.JournalofSafetyResearch

35,115–123.

Letirand,F.,Delhomme,P.,2005.Speedbehaviourasachoicebetweenobserving

andexceedingthespeedlimit.TransportationResearchPartF8,481–492. Long,J.S.,1997.RegressionModelsforCategoricalandLimitedDependentVariables.

SagePublications,ThousandOaks,CA.

Lund,A.K.,1993. Effectivenessofadministrativelicenserevocation(ALR)laws.

TransportationResearchBoardCircular,No.413TransportationResearchBoard,

(8)

Malenfant,J.E.L.,VanHouten,R.V.,Jonah,B.,2002.Astudytomeasuretheincidence ofdrivingundersuspensionintheGreaterMonctonarea.AccidentAnalysisand

Prevention34,439–447.

Manstead,A.S.R.,2000.Theroleofmoralnormintheattitude–behaviorrelation.In: Terry,D.J.,Hogg,M.A.(Eds.),Attitudes,behavior,andsocialcontext.Lawrence

ErlbaumAssociates,Mahwah,NJ,pp.11–30.

Marcil,I.,Bergeron,J.,Audet,T.,2001.Motivationalfactorsunderlyingtheintention todrinkanddriveinyoungmaledrivers.JournalofSafetyResearch32,363–376. Miles,D.E.,Johnson,G.L.,2003.Aggressivedrivingbehaviors:aretherepsychological

andattitudinalpredictors?TransportationResearchPartF6,147–161.

MoyanoDíaz,E.,2002.Theoryofplannedbehaviorandpedestrians’intentionsto

violatetrafﬁcregulations.TransportationResearchPartF5,169–175. Neyens,D.M.,Boyle,L.N.,2008.Theinﬂuenceofdriverdistractionontheseverity

ofinjuries.sustainedbyteenagedriversandtheirpassengers.AccidentAnalysis

andPrevention40,254–259.

NHTSA,1993.Policetimeandcostsassociatedwithadministrativelicense

revo-cation.ReportNo.HS-808064,U.S. Departmentof Transportation/National

HighwayTrafﬁcSafetyAdministration,Washington,DC.

Ouellette,J.A.,Wood,W.,1998.Habitandintentionineverydaylife:themultiple pro-cessesbywhichpastbehaviorpredictsfuturebehavior.PsychologicalBulletin 124(1),54–74.

Pai,C.W.,Saleh,W.,2008.Modellingmotorcyclistinjuryseveritybyvariouscrash typesatT-junctionsintheUK.SafetyScience46,1234–1247.

Parker,D.,Manstead,A.S.R.,Stradling,S.G.,Reason,J.T.,Baxter,J.S.,1992a.Intention tocommitdrivingviolations:anapplicationofthetheoryofplannedbehavior. JournalofAppliedPsychology77(1),94–101.

Parker,D.,Manstead,A.S.R.,Stradling,S.G.,Reason,J.T.,Baxter,J.S.,1992b.

Deter-minantsof intentionto commit drivingviolations. AccidentAnalysis and

Prevention24,117–131.

Parker,D.,Manstead,A.S.R.,Stradling,S.G.,Reason,J.T.,Baxter,J.S.,1992c. Inten-tiontocommitdrivingviolations:anapplicationoftheoryofplannedbehavior. JournalofAppliedPsychology77,94–101.

Parker,D.,Manstead,A.S.R.,Stradling,S.G.,1995.Extendingthetheoryofplanned

behaviour:theroleofpersonal norm.TheBritishPsychologicalSociety34,

127–137.

Parker,D.,Stradling,S.G.,Manstead,A.S.R.,1996.Modifyingbeliefsandattitudesto exceedingthespeedlimit:aninterventionstudybasedonthetheoryofplanned behaviour.JournalofAppliedSocialPsychology26,1–19.

Parker,D.,Lajunen,T.,Stradling,S.,1998.Attitudinalpredictorsofinterpersonally aggressiveviolationsontheroad.TransportationResearchPartF1,11–24. Pas,E.,1988.Weeklytravel-activity.Transportation15,89–119.

Poulter,D.R.,Chapman,P.,Bibby,P.A.,Clarke,D.D.,Crundall,D.,2008.Anapplication

ofthetheoryofplannedbehaviourtotruckdrivingbehaviourandcompliance

withregulations.AccidentAnalysisandPrevention40,2058–2064.

Ross,H.L.,Gonzales,P.,1988.Effectsoflicenserevocationondrunk-driving

offen-ders.AccidentAnalysisandPrevention20,291–379.

Schank,R.C.,Abelson,R.P.,1977.Scripts,plans,goals,andunderstanding.Erlbaum, Hillsdale,NJ.

Schank,R.C.,1982.Dynamicmemory.CambridgeUniversityPress,NewYork.

Sheehan,M.,Schonfeld,C.,Ballard,R.,Schoeld,F.,Najman,J.,Siskind,V.,1996.Athree

yearoutcomeevaluationofatheorybaseddrinkdrivingeducationprogram.

JournalofDrugEducation26,295–312.

S¸ims¸eko˘glu,Ö.,Lajunen,T.,2008.Socialpsychologyofseatbeltuse:acomparison oftheoryofplannedbehaviorandhealthbeliefmodel.TransportationResearch

PartF:Psychology11,181–191.

Siskind,V.,1996.Doeslicensedisqualiﬁcationreducereoffencerates?Accident

AnalysisandPrevention28(4),519–534.

Smith,D.I.,Maisey,G.,1990.Surveyofdrivingbydisqualiﬁedandsuspendeddrivers

inWesternAustralia.RoadSafetyTrends,3–4.WesternAustraliaDepartment

ofTransport,Perth,WA.

Sweedler,B.M.,Stewart,K.,1993.Reducingdrinkinganddrivingthrough

adminis-trativelicenserevocation.In:Proceedingofthe12thInternationalConference onAlcohol,DrugsandTrafﬁcSafety,Cologne.

Tertoolen,G.,VanKreveld,D.,Verstraten,B.,1998.Psychologicalresistanceagainst

attemptstoreduceprivatecaruse.TransportationResearchPartA32(3),

171–181.

Thuen,F.,Rise,J.,1994.Youngadolescents’intentiontouseseatbelts:theroleof attitudinalnormativebeliefs.HealthEducationResearch9,215–223. Triandis,H.C.,1977.InterpersonalBehavior.Brools/Cole,Monterey,CA.

Verplanken,B.,Aarts,H.,VanKnippenberg,A.,VanKnippenberg,C.,1994.Attitude

versusgeneralhabit:antecedentsoftravelmodechoice.JournalofApplied

SocialPsychology24,285–300.

Verplanken,B.,Aarts,H.,VanKnippenberg,A.,Moonen,A.,1998.Habitversus

plannedbehaviour:aﬁeldexperiment.BritishJournalofSocialPsychology37, 111–128.

Verplanken,B.,Aarts,H.,1999.Habit,attitude,andplannedbehaviour:ishabitan

emptyconstructoraninterestingcaseofgoal-directedautomatic?European

ReviewofSocialPsychology10,101–134.

Verplanken,B.,Orbel,S.,2003.Reﬂectionsonpastbehaviour:aself-reportindexof habitstrength.JournalofAppliedSocialPsychology33,1313–1330.

Victoir,A.,Eertmans,A.,VanDenBergh,O.,VanDenBroucke,S.,2005.Learningto drivesafely:social-cognitiveresponsesarepredictiveofperformanceratedby novicedriversandtheirinstructors.TransportationResearchPartF8,59–74. Voas,R.B.,DeYoung,D.J.,2002.Vehicleaction:effectivepolicyforcontrollingdrunk

andotherhigh-riskdrivers?AccidentAnalysisandPrevention34,263–270.

Warner,H.W.,Åberg,L.,2008.Drivers’beliefsaboutexceedingthespeedlimits.

TransportationResearchPartF11,376–389.

Webb,T.L.,Sheeran,P.,2006.DoesChangingBehavioralIntentionsEngender

Behav-iorChange? AMeta-Analysis of theExperimental Evidence. Psychological

Bulletin132(2),249–268.

Williams,R.L.,Hagen,R.E.,McConnell,E.J.,1984.Asurveyofsuspensionand revo-cationeffectsonthedrink-drivingoffender.AccidentAnalysisandPrevention

16(5–6),339–350.

Wood,W.,Quinn,J.M.,Kashy,D.,2002.Habitsineverydaylife:thought,emotion, andaction.JournalofPersonalityandSocialPsychology83,1281–1297. Wood,W.,Quinn,J.M.,2005.Habitsandthestructureofmotivationineverydaylife.

In:Williams,K.D.,Forgas,J.P.(Eds.),SocialMotivation:Consciousand Uncon-sciousProcesses.CambridgeUniversityPress,NewYork,pp.55–70.

Yagil,D.,1998.Genderandage-relateddifferencesinattitudestowardtrafﬁclaws

andtrafﬁcviolations.TransportationResearchPartF:TrafﬁcPsychologyand

Behaviour1(2),123–135.

Zador,P.,Lund,A.,Fields,M.,Weinberg,K.,1989.Fatalcrashinvolvementandlaws againstalcohol-impaireddriving.JournalofPublicHealthPolicy10,467–485.