Littering Behavior in America: 2009 Report

PreparedforKeepAmericaBeautiful

January,2009

LitteringBehaviorinAmericaResultsofaNationalStudy

910W.SanMarcosBlvd.,Ste.108SanMarcos,CA92078www.takeactionresearch.com

©2009KeepAmericaBeautiful

ReporttoKeepAmericaBeautiful i

TABLEOFCONTENTS

TABLEOFCONTENTS ...............................................................................................................................i

LISTOFFIGURES ...................................................................................................................................... iii

LISTOFTABLES ........................................................................................................................................ iv

ACKNOWLEDGEMENTS......................................................................................................................... vi

EXECUTIVESUMMARY ........................................................................................................................... 1

1.BEHAVIORALOBSERVATIONS................................................................................................... 1

2.INTERCEPTINTERVIEWS ............................................................................................................ 3

3.NATIONALTELEPHONESURVEY ............................................................................................. 4

CONCLUSIONSANDRECOMMENDATIONS............................................................................... 4

PARTI:INTRODUCTION ........................................................................................................................ 6

PROJECTGOALS..................................................................................................................................... 8

PARTII:BEHAVIORALOBSERVATIONS........................................................................................10

INTRODUCTION...................................................................................................................................10

METHODS...............................................................................................................................................10

RESULTS..................................................................................................................................................14

PARTIII:INTERCEPTINTERVIEWS................................................................................................29

INTRODUCTION...................................................................................................................................29

METHODS...............................................................................................................................................29

RESULTS..................................................................................................................................................30

PARTIV:TELEPHONESURVEY.........................................................................................................34

INTRODUCTION...................................................................................................................................34

METHODS...............................................................................................................................................35

ii LitterBehaviorinAmerica

RESULTS .................................................................................................................................................38

PARTV:CONCLUSIONSANDRECOMMENDATIONS...............................................................54

REFERENCES.............................................................................................................................................59

APPENDIXA:REFERENCETABLES ................................................................................................63

APPENDIXB:BEHAVIORALOBSERVATIONDOCUMENTS ..................................................67

GeneralLitterObservationBooklet............................................................................................68

CigaretteButtLitterObservationBooklet ...............................................................................71

APPENDIXC:INTERCEPTSURVEYINSTRUMENT...................................................................74

APPENDIXD:TELEPHONESURVEYINSTRUMENT.................................................................77

ReporttoKeepAmericaBeautiful iii

LISTOFFIGURES

Figure1:MapofSiteLocations .........................................................................................................11

Figure2:ObservedLitteringRatebyDistancetoReceptacle ..............................................21

Figure3:MeanLitterScoresforOverallLitterBehavior,LitterType,andLitterLocation ...................................................................................................................................39

Figure4:GroupMeansforLitteringBehavior(0‐10)byCurrentSmokerStatus.......42

Figure5:GroupMeansforLitteringBehavior(0‐10)byIndoorandOutdoorWorkEnvironment..........................................................................................................................43

Figure6:GroupMeansforLitteringBehavior(0‐10)byIndoorandOutdoorWorkEnvironmentsandSmoke‐FreeWorkplaces...........................................................44

Figure7:GroupMeansforCigaretteButtDisposalBehavior(0‐10)bySmoke‐FreeandNotSmoke‐FreeWorkplacesandOutdoorWorkEnvironment ............45

Figure8:GroupMeansforLitteringBehavior(0‐10)bySaworHeardLitterPreventionMessages .........................................................................................................48

iv LitterBehaviorinAmerica

LISTOFTABLES

Table1:TypesandNumberofTrashReceptaclesatLocations .........................................15

Table2.NumberofSiteswithExistingLitterbyType ...........................................................15

Table3.FrequenciesofWhereGeneralLitterwasLitteredinObservationLocations.....................................................................................................................................................16

Table4.FrequenciesofWhereCigaretteButtswereLitteredinObservationLocations.................................................................................................................................16

Table5.NumberofObservationsacrossSites ...........................................................................17

Table6.TypeandFrequencyofDisposedObjects ...................................................................18

Table7.FrequenciesofDisposalMethodsforGeneralLitter..............................................19

Table8.FrequenciesofLitterDisposalStrategiesforGeneralLitter ..............................20

Table9.FrequenciesofDisposalMethodsforCigaretteButtLitter.................................25

Table10.FrequenciesofLitterDisposalStrategiesforCigaretteButtLitter...............26

Table11.ReportedLocationofMessaging ..................................................................................30

Table12.ItemsReportedLittered...................................................................................................31

Table13.RespondentReportedLitteringLocations...............................................................31

Table14.PlaceofDisposalforCigaretteButtLitter................................................................32

Table15.Self‐ReportedLitteringBehaviorinthePastMonth ...........................................38

Table16.GroupMeansofLitteringBehaviorbyCommunityLivability.........................40

Table17.GroupMeansforLitteringBehavior(0‐10)byMotivatorsandBarriersItems .........................................................................................................................................46

Table18.GroupMeansofLitteringBehavior(0‐10)byGender,Education,TypeofResidence,andTypeofVehicle.....................................................................................49

Table19.UnstandardizedRegressionCoefficientsfromaMultipleRegressionAnalysis:LitteringBehaviorbyPredictorVariables ...........................................50

ReporttoKeepAmericaBeautiful v

Table20.GroupMeansforImportanceofLitteringinYourTown(0‐10)byCommunityLivabilityItems ...........................................................................................51

Table21.SitesbyLocationandObservationType...................................................................63

vi LitterBehaviorinAmerica

ACKNOWLEDGEMENTS

Contractor:ActionResearch,Inc.,910W.SanMarcosBlvd,Suite108,SanMarcos,CA,92078,USA.

PrincipalInvestigator:P.WesleySchultz,Ph.D.SeniorScientist,ActionResearchProfessorofPsychology,CaliforniaStateUniversity

ActionResearchTeam:LoriBrownLarge,M.A.ProjectManager,ActionResearch

JenniferTabanico,M.A.DirectorofResearch,ActionResearch

CoralBruni,M.A.DataManagementandAnalysis,ActionResearch

Collaborators:ReneeBator,Ph.D.AssociateProfessorofPsychology,SUNYPlattsburghActionResearchFieldTeam JennaAlbert JamieKuhnSaraAguilar NicoleLeFevreMichelleCugini AndreaMartinoTracyGalea EvaRichardsonElizabethMorales AshleeRockBelindaRojas AngelaBryan,Ph.D.(Coordinator,NM)MichaelStringham MaddieIkedaSUNYPlattsburghFieldTeam JennaKicklighterMontgomeryBopp StefanKlimajKaraCarpenter EvaPadillaAshleyDoyle JennaTonelliCassieFortney KianiWong

AspecialthankstoSusanneWoodsandthestaffatKeepAmericaBeautifulfortheirencouragement,support,anddedicationtothisimportantresearch.TheresearchsummarizedinthisreportwasmadepossiblewithfundingfromPhilipMorrisUSA,anAltriaCompany.

ReporttoKeepAmericaBeautiful 1

EXECUTIVESUMMARY

Likemanysocialproblems,litteriscausedbyhumanbehavior.Whetherintentionaloraccidental,litterbeginswiththeindividual.Giventhesocial,aesthetic,andenvironmentalproblemsthatresultfromlitter,numerousinterventionshavebeendeveloped,implemented,andevaluated.Yetdespitetheseefforts,littercontinuestobeaproblem.

Thisreportsummarizesthefindingsfromanationwidestudyoflitteringbehavior.Ourgoalswerethree‐fold:

• Toconductalandmarkstudythatcouldserveasaplatformforthenextgenerationoflitterpreventionactivities,

• Developavalidandreliablesetofmethodologiesthatcouldbereplicatedovertimeandindifferentlocations,

• Utilizeamulti‐methodapproach,includingbothself‐reportandbehavioralobservations,inwaysthatwouldallowforconclusionsaboutlitteratanationallevel.

Toachievethesegoals,theActionResearchteamconductedthreesetsof

studies,allwithnationwidesamples.Thefirstwasaseriesofbehavioralobservations,inwhichtrainedresearchersobservedthousandsofindividualsinadiversesampleoflocationsacrossthecountry.Thesecondsetofresearchactivitiesinvolvedinterceptinterviewswithasmallsampleoftheobservedindividuals.Finally,ourthirdresearchactivityinvolvedanationwidetelephonesurvey.InthisExecutiveSummary,wehighlightseveralofthekeyfindingsfromeachoftheseresearchapproaches.

1.BEHAVIORALOBSERVATIONS

Inanefforttogobeyondthetypicalself‐reportmeasuresusedtostudylitteringbehavior,ourresearchteamobservedindividualsinadiversesampleofpubliclocationsnationwide.Usingastrictresearchprotocol,theobserversrandomlyselectedindividualsin130locationsnationwide.Theserandomlyselectedindividualsweremonitoredastheymovedthroughthelocation,andtheirdisposalbehaviorswererecorded.

Inall,theteamobserved9,757individualsacross130differentlocations.Thelocationsweredividedacross10states,evenlysplitbetweenrural,urban,andsuburbansettings,andincludedninesitetypes:fastfood,recreation,gasstations,citycenters,reststops,medical/hospital,bars/restaurants,retail,andrecreation.

2 LitterBehaviorinAmerica

Themajorityoftheobservationsexaminedgeneraldisposalbehaviors,andasmallersubsetfocusedspecificallyonthedisposalofcigarettebutts.StatisticalanalyseswereconductedusingHierarchicalLinearModeling,whichallowedforanalysesofbothindividual‐andcontext‐levelpredictorsoflitteringbehavior.

KeyFindings:

• Trashreceptaclesarecommonatpublicplacesnationwide.Ofthe130locationsourteamvisited,118(91%)hadatleastonetrashreceptacle,includingtrashcan,tr/ashcombination,ashtray,dumpster,andrecyclingbin.Incomparison,cigarettereceptaclesarefarlesscommon,andonly61(47%)ofthesiteswevisitedhadatleastonereceptacle(ashtrayortrashcan/ashtraycombo).

• Despitethewidespreadavailabilityoftrashreceptaclesinpublicspaces,litterisstillquitecommon.Ofthe130siteswevisited,onlytwowerelitterfree.Themostcommonlyfoundexistinglitterwascigarettebutts(106sites,82%),miscellaneouspaper(87sites,67%),andfoodwrappers(58sites,45%).

• Commensuratewiththevolumeoflitter,ourteamobservedahighamountoflitteringbehavior.Ofallthedisposalsthatwewitnessedacrossoursample,17%werelitter.

• Themostfrequentlylittereditemwascigarettebutts(inourfocusedobservationsofsmokers,weobserveda65%litteringrate).Withregardtodisposals,ourteamalsoobservedhighlitteringratesforfoodremnantsandfoodwrappers.

• Contrarytoexpectations,themajorityoflitteringbehavior(81%)occurredwithnotableintent.Thisincludeddroppingwithnotableintent(54%),flick/flingoftheitem(20%),andothernotableintent(7%).

• Ourstatisticalanalysesrevealedthat15%ofthevarianceingenerallitteringbehaviorwasduetocontextualdemands,andtheremaining85%resultedfromtheindividual.Thatis,whilesometypesofcontextsinvitedmorelitter,therewasalargeamountofvariabilityinthebehaviorsofindividualswithinasite.Thisfindinghasimplicationsforprogramsdesignedtoreducelitteringbehavior:themosteffectivestrategywillincludebothcontextualefforts(likeclean‐ups,improvingwastecollectioninfrastructure,orbeautification)andeffortstoincreaseindividualmotivationtoproperlydispose.

• Withregardtogenerallittering,ourstatisticalanalysesrevealedseveralimportantpredictors.Attheindividuallevel,agewasstronglynegativelyrelatedtolitteringbehavior,witholderindividualslitteringlessthanyoungerindividuals.Atthelevelofthecontext,theavailabilityanddistanceoftrashreceptacleswasstronglypredictiveoflitteringbehavior.Inaddition,theamountoflitteralreadypresentcontributedtothelitteringrate.


• Withregardtocigarettebuttlitter,ourstatisticalanalysesshowedthatcigarettebuttlitterwasmorestronglyinfluencedbythecontextthanwasgenerallitter.Indeed,38%ofcigarettebuttlitteringwasduetocontext,and62%wasduetoindividualvariability.Attheindividuallevel,agewassignificantlypredictiveoflittering,witholdersmokersbeinglesslikelytolitter.Atthelevelofthecontext,boththeavailabilityofashreceptaclesandtheamountofexistinglitteraffectedthelitteringrate.Thisfindhasimportantimplicationsforprogramsdesignedtoreducecigarettebuttlitter:themosteffectivestrategywillinvolveincreasingtheavailabilityofashreceptacles,decreasetheamountofexistinglitterthroughclean‐upactivities,andmotivationalmessagestargetingindividualresponsibilityandobligation.

2.INTERCEPTINTERVIEWS

Inadditiontooursystematicobservationsofdisposalbehaviors,werandomlyselectedlocationswhereweconductedinterceptinterviewswiththeobserveddisposers.Ourgoalwastoobtainasmall,representativesampleofindividualsandtolinktheresponsesfromourinterceptinterviewtotheobserveddisposalbehaviors.Ateachoftheselectedinterceptlocations,individualswhohadbeenobserveddisposing(eitherproperlyorimproperly)wereapproachedtotakepartinaface‐to‐facesurvey.Theinterviewerdidnotknowifthepersonofinteresthadlitteredornot,andsheaskedasetofpre‐determinedquestions,includingdemographicitemsandquestionsaboutlitterbehavior.

Inall,102interceptinterviewswereconductedacross15locations.

KeyFindings:

• Consistentlyacrossthelocations,respondentsexpressedtoustheimportanceoflitteringasanissue,andtheyexpressedastrongpersonalobligationnottolitter.

• Asizeablepercentageofrespondentsreportedseeingorhearinglitterpreventionmessages(40%),messagesaboutcommunityclean‐upactivities(41%),ormessagesdiscouragingcigarettebuttlitter(25%).

• Yetwhenaskediftheyhadlitteredinthepastmonth,only43%ofrespondentsindicatedthattheyhadlittered.Interestingly,themostfrequentlymentioneditemsthatwerereportedmatchedthosethatweobservedinthelocations,andthosethatweobservedbeinglittered:cigarettebutts,foodwrappersorremnants,andpaper.

• Ourstatisticalanalysesshowedthatapersonalobligationnottolitterwassignificantlyrelatedtolowerratesofobservedlitteringbehavior.


3.NATIONALTELEPHONESURVEY

Finally,wecomparetheresultsfromoursystematicobservationstothosefromarandomdigitdialingtelephonesurvey.Dataarereportedonanationalprobabilitysampleof1,039respondents.Surveyitemsincludedself‐reportedlitteringbehavior,ratingsofcommunitylivability,personalnormsagainstlittering,attitudesaboutlitter,motivatorsandbarriersforlittering,anddemographicvariables.

KeyFindings:

• Usingthesameeightlitteringitemsreportedina1968nationalsurvey,wefoundthatonly15%ofthesamplereportedlitteringinthepastmonth.Thisfigureissubstantiallylowerthanthe50%admittedlitteringratefoundin1968,anditspeakstothedramaticdropinlitteringbehavioroverthepast40years.

• Acrossarangeofitemsandsituations,theoverallrateofreportedlitteringwasverysmall.Themostfrequentlyreportedlittereditemswerecigarettebutts,foodremnants(applecoreorbananapeel),andgum.

• Althoughthereportedlitteringratewassmall,itissignificantintermsofthenumberofpeopleitrepresentsandhowthatplaysoutinourcommunities.

• Communityappearancewasassociatedwithlowerratesofreportedlitteringbehaviors.Communityappearanceincludedcleanliness,lowratesofvisiblelitter,maintainedstreetsandsidewalks,attractivenessofplants,flowers,andtrees,andattractiveinfrastructure.

• Litteringwasreportedmorefrequentlyininstanceswhenthepersonwasinahurry,notrashcanwasnearby,theitemwasbiodegradable,therewasasensethatsomeoneelsewouldpickitup,andwhentheitemwasnotrecyclable.

CONCLUSIONSANDRECOMMENDATIONS

Takentogether,theresultsprovideaglimpseintolitteringinAmerica.Whilelitteringrateshavefallensubstantiallyoverthepast40years,litteringremainsanimportantsocialandenvironmentalissue.Themethodologydevelopedforthisprojectwasintendedtoinformnational‐levellitterpreventionstrategies.However,theapproachalsolendsitselftounderstandingtheissuespertinentatthecommunitylevel.Themethodologyandprotocolswerecreatedinsuchaway(and


reportedinsufficientdetail)toallowforreplicationinlocalcommunities,andtoprovideacoresetofmeasurestoassesschangesinlitteringratesovertime.


PARTI:INTRODUCTION

Likemanyenvironmentalproblems,litteriscausedbyhumanbehavior.Whetherintentionaloraccidental,litterbeginswiththeindividual.Giventhesocial,aesthetic,andenvironmentalproblemsthatresultfromlitter,numerousinterventionshavebeendeveloped,implemented,andevaluated.Unfortunately,themajorityoftheseprogramsarenotbasedonsoundprinciplesofhumanbehavior.Whilethereisanexistingresearchbaseonlitterdatingbackmorethan40years,thefindingshavenotbeensystematicallyintegratedandthetopicsstudiedareonlylooselyconnected.

Inresearchconductedinthe1970s,KeepAmericaBeautiful(KAB)foundthatlitteroriginatedfromsevenprimarysources:(1)pedestriansorcyclistswhodonotusereceptacles,(2)motoristswhodonotusecarashtraysorlitterbags,(3)businessdumpstersthatareimproperlycovered,(4)loadingdocksandcommercialorrecreationalmarinaswithinadequatewastereceptacles,(5)constructionanddemolitionsiteswithouttarpsandreceptaclestocontaindebrisandwaste,(6)truckswithuncoveredloadsonlocalroadsandhighways,and(7)householdtrashscatteredbeforeorduringcollection(KAB,2007).

Thefindingsfromthisearlyresearchservedasafoundationtoinformoutreachandclean‐upactivities.Sincethattime,therehavebeenmorethan100studiesconductedonthetopicofpubliclitter.Thesestudieshavebeenconductedbyprivateresearchfirms,nongovernmentalorganizations,academicresearchgroups,andtoalesserextent,localentities(e.g.,cities,counties).Themostfrequentlystudiedtopicsinclude:littercomposition,wherelitteroccurs,wholitters,andlitterpreventionstrategies.Whilethecurrentprojectwasdesignedtoupdateandinformthenextgenerationoflitter‐preventionactivities,itisusefultobeginwithashortsummaryoftheexistingknowledge.

1.WhatisLitter?AccordingtoGeller(1978),litterconsistsofitemsfoundinsociallyunacceptablelocations.However,itisalsoa“functionofspecificityandconvenience.”Studieshaveshownthatthetopfivetypesoflitterare:(1)miscellaneouspaper,(2)miscellaneousplastic,(3)vehicledebris,(4)packaging,and(5)beveragecontainers(R.W.Beck,2007).Arecentreviewofstudiesonlitter,from1993to2006,foundthatthetop10mostfrequentlylittereditemswere:(1)takeoutfoodpackaging,(2)snackwrappers,(3)miscellaneouspaper,(4)miscellaneousplastic,(5)vehicledebris,(6)beveragecontainers,(7)napkins,bags,andtissues,(8)miscellaneousmetalandglass,(9)otherbeveragerelatedlitter,and(10)constructiondebris.Thesestudiesdidnottakeintoaccountcigarettebuttsasaformoflitter,andotherstudieshaveshownthatcigarettebuttsarethemostfrequentlylittereditemwhencounted,andtheyarethenumberoneitemlitteredbyvolume(DepartmentofEnvironmentandConservation,2004;Sibley&Liu,2003).


2.WhoLitters?Recentstudieslookingatthetypesoflitteringbehaviorovertime(from1990topresent)haveshownthatthereisatrendawayfromdeliberatelittering,withlitteringasawholedroppingbyapproximately2%peryearsincethe1990s(R.W.Beck,2007).Youngerpeoplearemorelikelytolitterthanolderpeople,withthoseaged19andbelowmorelikelytolitterthananyotheragegroup(Krauss,Freedman,&Whitcup,1978;RWBeck,2007).Onestudysuggestedthatadults21‐35arethreetimesmorelikelytolitterthanthoseovertheageof50andtwotimesmorelikelytolitterthanthoseaged35‐49(Geller,1968).

Otherdemographicpredictorsoflitteringincludegender,urbanliving,andhouseholdsize.Whilegendereffectsinlitteringtendtobesmall,menaremorelikelytoadmitlitteringthanwomen(Cialdini,Reno,&Kallgren,1990;Kraussetal.,1978;Meeker,1997).Inaddition,thereissomeevidencethatpeoplelivinginruralareasaremorelikelytolitterthanpeoplelivingincities.Andindividualsfromsmallhouseholds(1‐2)arelesslikelytolitterthanthosefromlargehouseholds(5+)(Geller,1968;ascitedbyR.W.Beck,2007).

3.PreventingLitter.Athirdareaofresearchonlitterhasfocusedonstrategiesforreducinglitter.

A.Litterbegetslitter.Oneofthelargestfactorsaffectingaperson’sdecisiontolitteristheconditionofthephysicalsurroundings.Individualsaresubstantiallymorelikelytolitterintodirtyoralreadylitteredenvironmentsthanintocleanones(Cialdini,Kallgren,&Reno,1991;Geller,Witmer,&Tuso,1977;Herberlein,1971;Reiter&Samual,1980).Thedifferenceisoftenasmuchas2‐3timesasmuchlitterindirtyenvironments.

B.Litteringfollowsthenorm.Numerousstudieshaveshownthatsocialnormsplayapowerfulroleinanindividual’sdecisiontolitter.Surveydatasuggeststhatpersonalnormsaboutlitteringhavechangedconsiderablyoverthelast50years,fromamoderatelevelofconcernforlitteringinthe1950s,toastrongfeelingofpersonalobligationtonotlitterbytheearly2000s(cf.Grasmick,Bursik,&Kinsey,1991;Kallgren,Reno,&Cialdini,2000).Associalsanctionsagainstlitteringgrew,studiesshowedthatactivatingnormsaboutlitteringcouldsubstantiallyincreaseordecreaseone’sownlitteringdecisions.Forexample,seeingapersonlitter,seeingapileofsweptlitter,orseeingapersonorgrouppickuplitter,canallactivatenormsandmobilizebehavior.Thesesituationsactivateaperson’snormativebeliefs,andtheresultingfeelingsofshame(violatingasocialnorm)andguilt(violatingapersonalnorm)havebeenincorporatedintoPublicServiceAnnouncementswithdocumentedsuccess(Cialdini,2003;Cialdinietal.,2005).


C.Theconvenienttruth:Athirdestablishedfindingisthatlitteringratesdecreaseastheconvenienceofusingaproperreceptacleincreases(Cope,Huffman,Alred,&Grossnickle,1993;Geller,Brasted,&Mann,1980;Geller,Winett,&Everett,1982).Asageneralrule,moretrashreceptaclesreducelitter,includingcigarettebutts.Furthermore,thesalienceofthereceptaclecanalsodecreaselitterrates.Abrightlycolored,themed,ordecoratedreceptacleattractsconsiderablymoretrashthandoesaplainorordinaryappearingreceptacle.

D.Programs:ProgramsthroughorganizationssuchasKABandAdopt‐a‐Highwayhaveworkedtoreducelitter.StudieshaveshownthatsiteswhereKABandAdopt‐a‐Highwayprogramswerepresentwerecleanerandhadlesslitterthansiteswheretheseprogramswerenotpresent(R.W.Beck,2007).

E.Sanctions.Whilethreatsoffinesandpunishmentarecommonplaceinlitter

preventionprograms,therearefewsystematicevaluationsoftheirefficacy.Infact,researchbySansoneandHarackiewicz(2000)suggeststhatthreatcanundermineintrinsicmotivation(e.g.,themotivationtodosomethingbasedonenjoyment).Asaresult,anychangeinbehaviorthatresultsfromthreatsorsanctionsislikelytobespecifictothebehaviorathand,limitedtoinstanceswheresuchthreatsofpunishmentareenforceable,limitedintheabilitytogeneratelong‐termresults,andnotgeneralizedacrosssituations.Inaddition,thewayathreatofenforcementiswordedmayalsoplayapartinwhetherornotitiseffective,suchthatjustmentioningtheundesirablebehaviorcanactasaprime.Recentpsychologicalresearchhassuggestedthatprimingabehavioralcategoryincreasestheprobabilitythatitwilloccur(Aarts&Dijksterhuis,2003;Bargh&Chartrand,1999).Thus,enforcementmaynotonlybespecifictotheextrinsicmotivationsmentionedabove,butitmayalsotriggerthewrongeffectandproducemorelitteringinareaswhereenforcementisused.

PROJECTGOALS

Itisagainstthisbackdropofpriorworkthatwedevelopedourcurrentresearchplan.Ourgoalswerethree‐fold:

• Assembleanaccomplishedteamtodesignandimplementalandmarkstudythatcouldserveaplatformforthenextgenerationoflitterpreventionactivities,

• Developavalidandreliablesetofmethodologiesthatcouldbereplicatedovertimeandindifferentlocations,

• Utilizeamulti‐methodapproach,includingbothself‐reportandbehavioralobservations,inwaysthatwouldallowforconclusionsaboutlitteratanationallevel.


TheActionResearchteamsetouttoconductanextensiveandrigoroussetofstudiesexamininglitteringbehavior.Whilethetraditionalapproachtolitterresearchhasfocusedonthelittereditemsthataccumulatealongroadwaysorinpublicspaces,ouremphasiswasonthebehavioritself.Thatis,ourfocusinthisworkisonunderstandingthefactorsthatcontributetolitteringbehavior,andnotonthelitterperse.Webuiltourresearchdesignandanalysisalongthepremisethatlitterresultsfromhumanbehaviorandthatunderstandingthisbehavioriscentraltoanyefforttopreventandreducelitter.Thereportbelowisdividedintothreesections,eachsummarizingthemethodsandresultsfromdifferentaspectsofourresearchproject.

Inthefirstsection,wereportfindingsfromsystematicobservationsoflitteringbehavior.Theseobservationsaretheheartofourapproachandthemethodologyallowsustogobeyondthetypicalsurveyapproachandtoobservethebehaviorsofindividualsacrossarangeofeverydaysettings.Theanalyseswereportexamineboththeindividual‐levelandlocation‐levelpredictorsoflitteringbehavior.Ontheonehand,iflitteringresultsfromindividual‐levelvariableslikelackofconcern,apathy,orlowmotivation,thenwewouldexpecttoseeconsiderablevariabilitywithinlocations,andlittlevariabilityacrosslocations.Butontheotherhand,litteringbehaviormayresultfromcontextualvariables(location‐level)likelackofreceptacles,nosignage,ortheculturalmilieuofaparticularcityorregionofthecountry.Themethodologyandanalysesreportedbelowtakeafirststeptowardansweringthisimportantquestion.


PARTII:BEHAVIORALOBSERVATIONS

INTRODUCTION

BetweenMayandJuly,2008,ourresearchteamconductedsystematicobservationsofindividualsinawiderangeofpubliclocationsacrossthecountry.Theresearchdesignwasdevelopedasamulti‐levelmodel,andanalyzedusingHierarchicalLinearModeling(HLM)statisticalsoftware.Thenumberoflocations,numberofsites,andnumberofindividualswereselectedtooptimizethestatisticalpowerandgeneralizabilityoftheresults.Ateachlocation,werandomlysampledindividualsandthenmonitoredtheirbehaviorastheymovedthroughthesite.Separateobservationprotocolsweredevelopedformonitoringthebehaviorofsmokersandforobservationsofgenerallitteringbehaviors.

METHODS

SiteSelection

Thespecificsitesselectedfortheobservationalresearchcomponentweretheresultofastratifiedsamplingprocedure.Webeganbysamplingregionsofthecountry,states,cities,typesofsites,andfinallyspecificlocationsandaddresses.

State:Atotalof10stateswereselectedfromacrossthecontinentalUnitedStates:Arkansas,California,Georgia,Illinois,Kentucky,Nevada,NewMexico,NewYork,Utah,andVermont.Thestateswereselectedtorepresentavarietyofregions,aswellasstateswithvaryinglevelsoftobaccouse(seeFigure1).AdulttobaccouseprevalencedatawasgatheredfromtheCentersforDiseaseControlandPrevention(http://apps.nccd.cdc.gov/brfss)andwesampledbothhigh‐prevalence(e.g.,Kentucky)andlow‐prevalence(e.g.,Utah)states.Thefinalselectionofstatesrepresentedadiverserangeofthecountry’spopulation.


Figure1:MapofSiteLocations

City:Atotalof30citieswereselectedfromacrossthe10states.Threecitieswereselectedineachstateinordertorepresenturban,suburban,andruralareas.TheselectionofthethreecitieswithineachstatewasmadebaseduponpopulationdensitydataobtainedfromtheU.S.Census,proximitytoprimarytransportationpoints,aswellaslogisticalandsituationalconstraints(e.g.,distancebetweencities).SeeTable21inAppendixAforacompletelistofstates,cities,andsites.

Sites:Therewerefivesitetypesforthegenerallitterobservations(fastfood,recreation,gasstations,citycenters,andreststops),andfoursitetypesforthefocusedcigarettelitterobservations(medical/hospital,bars/restaurants,retail,andrecreation).ThesitetypeswerechosenincooperationwithKABtorepresentcriticallittersourceareas,aswellasfortheirpotentialtoprovideopportunitiesforstudyinghumanbehavior.Attheinitialselection,eachcityhadanequalopportunityofbeingassignedtoanyonesitetype.Forboththecigarettelitterandgenerallitterobservationsites,afullycrosseddesignwasusedtorepresentallofthepossiblecombinationsofcityandsite,andeachpossiblecombinationwastreatedasasite.Thetargetresearchsiteswerethenrandomlyselectedfromthelistofallpossiblecombinations.Aninitialtargetof70siteswasselectedforgeneral


litterobservations,and30sitesforcigarettelitterobservations.Duetologisticsandpracticalconsiderations,thefinaldatasetincluded130locationsacrosstheU.S.:86generallitterand44focusedoncigarettedisposals.Ofthetotalnumberofsites,therewere30recreationallocations,24citycenters,22fastfoodrestaurants,12retailstores,12barsorrestaurants,11gasstationswithconveniencestores,11reststops,and8medicalhospitals.

Locations:Withtheexceptionofcitycentersandreststops,theexactsitelocationswerechosenusinginformationfromwww.areaguides.netandarandomselectionmethod.Forexample,forgasstationsites,alistofallretailgaslocationsintheselectedcitywasgenerated,andfromthatlistthesitewaschosenrandomly(alongwithseveralback‐ups).Usingarandomselectionprocedureatthesitelocationleveleliminatedpotentialbiasbyensuringthatlocationswerenotselectedbecauseofconvenienceoreaseofaccessbythefieldresearchteam.Alllocationswereselectedpriortotravelingtotheobservationarea.Thefieldresearchteamtraveledwiththerandomizedlistofalllocationstoallowthemtomovetothenextlocationonthelistifforsomereasonthefirstlocationdidnotmeetthecriteriaspecifiedintheobservationguidelines.Observationswereconductingduringthedayandeveninghours.

ObservationProtocol

Systematicobservationsweremadeoftheindividualsineachsitefollowingastrictresearchprotocol.Theprotocolincludeddetailedproceduresfordefiningthephysicalboundariesoftheobservationarea,randomselectionofindividuals,locationdescriptioninformation,andbehavioralobservations.AppendixBprovidesamoredetaileddescriptionofthematerialsused,andthebasicprotocolisoutlinedbelow.

1. DefiningofPhysicalBoundaries:Oncethefieldresearchteamarrivedattheselectedlocation,thephysicalboundariesoftheobservationareaweredefined.Individualswererandomlysampledandobservedwithinthedefinedphysicalboundaries.Oncetheindividualexitedthisboundary,theyweresaidtohave“leftsite”andtheobservationwasterminated.

2. LocationDescriptionWorksheets:Locationdescriptionworksheetswerecompletedinordertocharacterizethephysicalsurrounding,andtoallowforanalysesofthedegreetowhichparticularsitecharacteristicsinfluencedthedisposalbehaviors.Includedinthisworksheetweremeasurementsof:

• PhysicalAspects:Theseincludedweather,temperature,timeofobservationperiod,informationaboutthepresenceandphysicalcharacteristicsoftrashreceptacles(includinghowfullandtype),recyclingbins,andashreceptaclesatthelocation.Inaddition,


informationaboutthepresenceoflitterandcigarettebuttsatagivenlocation,andthepresenceofplantersandtheirlitteredconditionwasalsonoted.Attheendofeachobservationperiod,observerscodedthehighestandlowestnumberofpeopleatthesite,andminimumandmaximumcrowdednessduringtheobservationperiod.

• Livability:Animportantaspectofthelocationdescriptionworksheetwasthesectiononthephysicalappearanceofthesite.Inthiswecompiledalivabilityquotient,whichincludedinformationaboutthesite’scleanliness,walkability,landscaping,andinfrastructure.Eachaspectwasratedonascaleof0to10,with10beingthebestinthatcategory.

• SignageandEnforcement:Informationregardingthepresenceofanti‐litteringandanti‐smokingsignageandenforcementswerenoted.

• SketchandPictures:Asketchofeachlocationwascreatedandpicturesweretakenofeachlocation.PicturesofeachlocationareprovidedonaCDaccompanyingthisreport.

Priortoconductingourobservations,theteamunderwentextensivetraining,andavarietyofstrategieswereusedtoensurereliableandvaliddatacoding.

3. BehavioralObservations:Behavioralobservationswereconductedateachsiteduringtheobservationperiod.Observationsgenerallybeganinthemorning,andlasteduntilthedayended,oruntilthetargetnumberofobservationswerereachedatthesite.Observationsincluded:

• GeneralLitter:Allagesofindividualswereobservedinthegenerallitterobservations.Individualswereobserveduntiltheydisposedofanitemorlefttheobservationfield.Duringtheseobservations,wecodedwhethertheindividualdisposedofanitem,andifsowhetherthedisposalwasproperorimproper.Duringtheobservations,notationswerealsomaderegardingdemographiccharacteristicsoftheindividuals(e.g.,approximateageandgender).Otherindividual‐levelcharacteristicssuchaspresenceofothergroupmembersandproximitytotrashreceptacleswerealsocoded.

• CigaretteButtLitter:Onlysmokersovertheageof21wereincludedintheobservations.Smokerswereobservedinordertobetterunderstandhowoftenandunderwhatconditionstheylittercigarettebutts.Smokerswereobserveduntilthecigarettewasdisposedortheylefttheobservationfieldwiththecigarette.Individualandgroupdynamicswererecorded.Fromtheseobservations,disposalswerecodedasproperorimproper.Duringtheobservations,notationsweremaderegardingtheirapproximateageandgender.


4. Randomization:Throughouttheobservationalperiod,researchersassessedtheflowoftraffic(i.e.,numberofpeople)atthesiteandchoseanappropriaterandomizationsequencetoobtainarepresentativesampleofindividuals.RandomizationwasachievedbyselectingtheNthperson,withNbasedontheflowoftrafficandrangedfromobservingeveryperson(N=1)toobservingeverysixthperson(N=6).Randomselectionofindividualsateachsiteisakeyaspectofourresearchprotocol,andonethatallowsustocalculatealitteringrateforeachsite,andtoestimatethelitteringratenationally.1Thestrengthofthiselementshouldnotbeunderestimated,andteamsconsideringasimilarstudyoflitteringbehaviorarestronglyadvisedtoimplementrandomsamplingtechniques(ratherthansimplyfindingandrecording“disposers”).

RESULTS

ObservationSites

Theobservationsitesrepresentalargeanddiversesampleoflocationsacrossthecountry.Observationswereconductedat130locationsacrossthecountry:86generallitterand44observationsofsmokers.Ofthetotalnumberofsites,therewere30recreationallocations,24citycenters,22fastfoodrestaurants,12retailstores,12barorrestaurants,11gasstationswithconveniencestores,11reststops,and8medicalhospitals.Notethatthisnumberexceededourtargetsampleof100foravarietyofreasons(e.g.,insomelocationsthetargetof30observationswasreachedquicklyandtheteamwasabletovisitasecondsiteinthesamecity,insomelocationstheflowoftrafficwastooslowandafterafewobservationstheteamchosetomovetothenextrandomlyselectedsite).Fortysiteswererural,45sitesweresuburban,and45siteswereurban.Theweatherattheselocationsduringtheobservationswasmostlysunny(34%),orcloudy(partlycloudy=26%;mostlycloudy=14%,andcloudy=16%),withthemeantemperatureof71degreesF.Only4%ofthelocationshadsomeformofrain(lightshowers=2%;heavyshowers=2%).At3%ofthelocationsitwaswindyandat1%ofthelocationstherewaslightsnow.

At118oftheselocations(91%)therewasatleastonetrashreceptacle(trashcan,trashcan/ashtraycombination,ashtray,dumpster,recyclingbin).Onaveragethereceptacleswere45feetapart(SD=44.22).Table1showsthetypeandnumberoftrashreceptaclesnotedatthelocations.Becausemostsitescontainedmorethanonetypeofreceptacle,thetotalsumstomorethan130.1Notethatourestimatesfornational‐levellitteringratesmustbequalifiedbythefactthatwedidnotrandomlyselectlocationsacrossthecountry.Thusourestimatedlitteringratesapplyonlytotheeighttypesoflocationsweobserved.


Table1:TypesandNumberofTrashReceptaclesatLocations

Numberofreceptaclesatlocation

1 2 3 4 5+ TotalSites

TrashCan,Uncovered(nolid) 18 17 11 8 10 64

TrashCan,Covered(withlid) 22 5 6 8 17 58

TrashCan/AshtrayCombo 6 6 1 1 4 18

AshtrayOnly 24 11 3 2 3 43

Dumpster(s) 8 4 ‐ ‐ ‐ 12

RecyclingBin(s) 3 9 1 3 ‐ 16

Theresearchteamratedtheextenttowhicheachlocationwaslittered,onascalefrom0(notatalllittered)to10(extremelylittered).TheexistingKABLitterIndexusesa4‐pointscale,butourgoalherewastoallowforamorerefinedanalysis.Theteamalsocountedandcodedthetypeoflitterineachlocation.Onlytwositeshadnolitter.Thetopitemsfoundwere:cigarettebutts(N=106sites),paper(N=87sites),andfoodwrappers(N=58sites).Table2showsthenumberofsiteswithvarioustypesoflitter.Table3showsthelocationofthelitterateachlocation.Again,thetotalexceeds130becausemanyofthesitescontainedmorethanonetypeoflitter.

Table2.NumberofSiteswithExistingLitterbyType

Frequency Percentage Cigarette Butts 106 82% Paper 87 67% Food Wrappers 58 45% Confections 44 34% Napkin/Tissue 44 34% Misc. Plastic 43 33% Food Remnants 31 24% Beverage Cup 21 16% Beverage Bottle: Plastic 14 11% Food Containers 12 9% Plastic Bags 11 8% Beverage Can 8 6% Beverage Bottle: Glass 6 5% Yard Waste 6 5% Other 35 27%


Table3.FrequenciesofWhereGeneralLitterwasLitteredinObservationLocations

Sites PercentageGround 121 93%Bushes/Shrubbery 49 38%Onoraroundtrashreceptacles 31 24%Planters 19 15%Other 10 8%

Theresearchteamalsoratedandcountedthenumberofcigarettebutts.Whenratedonascaleof0(notatalllittered)to10(extremelylittered),theaveragesitewasratedat3.48(SD=2.61).Whenlitteredcigarettebuttswerecounted,theaveragesitehad69cigarettebutts(SD=138)withintheobservationfield,withonesitehavingmorethan1,000litteredcigarettebutts.Whenexaminingwherethecigarettebuttswerelitteredintheselocations,110locationshadcigarettebuttslitteredontheground.Table4showsthefrequencyofwherethecigarettebuttswerelitteredattheobservationlocations.

Table4.FrequenciesofWhereCigaretteButtswereLitteredinObservationLocations

Frequency PercentageGround 110 85%Bushes/Shrubbery 48 37%Onoraroundtrashreceptacles 32 25%Planters 20 15%Other 6 5%

Thenumberofplantersinasiterangedfrom0to40,withtheaveragesitehaving2.10plantersinitslocation.Whenthetypeoflitterthatwasintheseplanterswasnoted,10siteshadnolitterintheirplanters.However,themostfrequenttypeoflitterintheplanterswascigarettebutts,with21locationshavingplanterslitteredwithcigarettebutts.

Thephysicalappearanceofeachlocationwasalsomeasured.Usingacompiledmetric,sitesrangedfrom2.5to10onascalefrom0(nobeautification)to10(extremebeautification),withtheaveragesiteratingof6.41(SD=1.76).

Thenumberofpeopleinthelocationswasalsonotedatboththelowestandhighestpointsduringtheobservationperiod.Crowdednesswasalsoratedonascaleof0(notatallcrowded)to10(extremelycrowded).Atthebeginningoftheobservationperiod,theaveragecrowdednessofalocationwas1.21(SD=1.73)andatthepeak,theaveragecrowdednessofalocationwas4.99(SD=2.39).


GeneralLitterObservations

Observationsweremadeat86sites,across10states.Ofthese,therewere23citycenters,22fastfoodrestaurants,19recreationalsites,11reststops,and11gasstationswithconveniencestores.Atotalof8,990observationsweremade,with33%fromruralareas,34%fromsuburbanareas,and33%fromurbanareas.Twenty‐sevenpercentoftheobservationsweremadeinthemorningbefore11:59a.m.,58%weremadeintheafternoonbetweenthehoursof12:00p.m.and3:59p.m.,and16%weremadeintheeveningsbetweenthehoursof4:00p.m.and7:59p.m.ThenumberofsitesandobservationsisshowninTable5.

Table5.NumberofObservationsacrossSites

General Sites

General Observations

Cigarette Sites

Cigarette Observations

Recreational Sites 19 1206 11 82 City Center 23 2648 1 30 Fast Food 22 2294 Bars and Restaurants

12 97

Retail 12 354 Rest Stops 11 1398 Gas Stations 11 1444 Medical/Hospital 8 204 TOTALS 86

Sites 8990

Observations 44 Sites 767

Observations

Fifty‐sixpercentofthoseobservedweremaleand44%werefemale.Observedagesrangedfrom1to82years(M=37.92,SD=15.97),and50%ofthoseobservedwereinagroup.

Ofthe8,990peoplewhowereobserved,2,472leftthesitewithnoobject(28%),4,534leftsitewithanobject(50%),and1,962disposedofanobject(22%)whileonsite.Twentytwooftheobservationswerenotcodable(e.g.,theobserverwasunabletoseethebehavior).Acrossthefullnumberofgeneralobservations,therewere342observedinstancesoflittering.Ofallindividualsobserved(N=8,990)intheselectedlocations,4%littered.Amongthesampleofindividualswhodisposedofanitem(N=1,962),17%disposedimproperly.

Disposals.Amongthe1,962observeddisposalbehaviors,themostfrequentlydisposeditemswere:cigarettebutts(N=340),combo/mixedtrash(N=337),andpaper(N=271).Table6showsthetypesandfrequenciesoftheobjectsdisposed.Intermsofpercentage,cigarettebutts(57%),foodremnants(20%),andfoodwrappers(14%)werethemostfrequentlyimproperlydisposed.Includedinthe


“other”categorywereseverallowfrequencydisposals,including:petwaste,confections,matches,diapers,straws,chewingtobacco,andproductpackaging.

Table6.TypeandFrequencyofDisposedObjects

Proper Improper % Littered Cigarette Butt 146 194 57% Combo/Mixed Trash 325 12 4% Paper 251 20 7% Beverage Cup 180 5 3% Napkin/Tissue 110 9 8% Beverage Bottle: Plastic 100 5 5% Food Remnants 65 16 20% Food Wrapper 85 14 14% Beverage Can 59 8 12% Food Container 57 1 2% Plastic Bag 38 2 5% Beverage Bottle: Glass 11 0 0% Unknown 116 10 8% Other 77 46 37% TOTAL 1,620 342 17%

Withregardtomethodofdisposal,usingatrashreceptaclewasthedominantmethod(N=1180),followedbylitteringontheground(N=287),pocketing(N=176),andhandingtheitemofftosomeoneelse(N=120).ThecategorizationofalldisposalsisshowninTable7,alongwiththefrequencyandpercentageofdisposals.


Table7.FrequenciesofDisposalMethodsforGeneralLitter

Frequency PercentageTrashReceptacle 1180 60%Ground† 287 15%Pocketed 176 9%Handoff 120 6%Ashtray 113 6%RecyclingBin:Correct 20 1%On/AroundTrashReceptacle† 14 1%LeftonTable,Bench,orLedge† 17 1%Bushes/Shrubbery† 10 1%Otherimproper† 4 <1%Planter† 8 <1%Separated(trash/recycle) 7 <1%Waterway(riverorlake)† 2 <1%RecyclingBin:Incorrect 2 <1%Otherproper 2 <1%TOTAL 1962 100%†Designatesmethodscodedasimproper

Littering.Ofthe1,962observeddisposals,342werecodedasactsoflittering.Eachactoflitteringwasclassifiedastothetypeofmaterialandtothestrategyoflitter.Themostfrequentlylittereditemswerecigarettebutts(N=198),followedbypaperornapkins(41and14respectively,N=55),foodremnantsorwrappers(N=39),andbeveragecupsorcans(N=37).

Thelitterstrategywascodedusinganadaptationofthecategoriesdeveloped

byapriorstudyinAustralia(Williams,Curnow,&Streker,1997):• Dropwithintent—thepersonsubtlydroppedtheitemontheground,

butwithnotableintenttolitter.Wedubbedthisgroup“DirtyDroppers.”

• Flagrantflingers:blatantlitteringofobjectsoutintheopenwithoutattempttoconcealtheiractions.

• Inchaway:“inchersplacetheirlitterbesidethemandslowlyinchawayfromituntiltheynolongerbeidentifiedasthelitterer.”

• Foulshootersaimtodiscardtheiriteminareceptacle,butaftermissingleavetheobjectontheground.

• Wedgersareindividualswho“stuffdisposableobjectsintosmallspaceswheretheywillnotbeseen.

• Sweeperscollecttheirlitterbybrushingitoffaflatsurfaceintotheirhandorabag,butintheprocessdropitemsontheground.


• 90%ersdisposeofthelargeconspicuousitems,butleavebehindaslitterthesmallerlessvisibleobjects.

Whenitemswerelittered,dropwithintentwasthemostfrequentlyused

strategy(N=183,54%),followedbyflagrantflicking(N=68,20%),and12%dropbyaccident.Dropwithintentincludesspitting,confections,anditemsintentionallyleftbehind.Dropwithoutintentincludesitemsthatleftbehindwithnonoticeableintent,anditemsthatthepersondroppedwithoutanyobservedrecognition.Combined,wefindthat81%ofobservedlitteringactsoccurredwithnotableintent.Table8showsthefrequencyoflitterstrategiesused.

Table8.FrequenciesofLitterDisposalStrategiesforGeneralLitter

Frequency PercentageDrop:Intent† 183 54%Flick/Fling† 68 20%Drop:NoIntent 42 12%Shoot&Miss† 8 2%Placeditemontable,bench,orledge† 21 6%InchAway† 8 2%Wedge† 4 1%Sweep† 3 <1%90%ers† 2 <1%Blewaway 3 <1%TOTAL 342 100%Note:†designateslitteringthatoccurredwithnotable“intent.”

Theobservationteamalsocodedthedistance(infeet)fromthedisposertothenearestreceptacle(trash,recycling,orashtray).Whiletherewereseveralinstancesoflitteringthatoccurredimmediatelyadjacenttoareceptacle,mostlitteringoccurredataconsiderabledistancefromareceptacle.Atthetimeofimproperdisposal,theaverageestimateddistancetothenearestreceptaclewas29feet.AsshowninFigure2(below),theobservedlitteringratewhenareceptaclewas10feetorcloserwasonly12%,andthelikelihoodoflitteringincreasedsteadilyforreceptaclesatagreaterdistance.Forreceptaclesthatweremorethan60feetaway,thelitteringrateremainedrelativelystableat30%.


Figure2:ObservedLitteringRatebyDistancetoReceptacle

UnderstandingGeneralLitteringBehavior

Aseriesofstatisticalanalyseswereconductedtoexaminetheindividualandcontextualvariablesthatarepredictiveoflittering.TheprimaryanalysiswasconductedusingHierarchicalLinearModeling(HLM),inwhichtheindividualwastreatedasLevel1(gender,age,inagroup),andthesitecharacteristicswereanalyzedasLevel2(e.g.,presenceofreceptacles,presenceoflitter,physicalappearance,rural/urban/suburbanlocation,signage/litterenforcement).Theanalysiswasconductedusingonlydatafromobservationswhereadisposaloccurred(N=1,962).

Theinitial“null”modelshowedthattheoveralllitterratewas.17.Acrossthe1,962individuals(σ=.12,Z=30.66,p<.01)andthe86locations(τ00=.022,Z=5.71,p<.01),therewasconsiderablevariabilityinthelitterrate.TheIntraClassCorrelationcoefficient(ICC)was.15.Thisstatisticisdirectlyinterpretableanditindicatesthat15%ofthevarianceinlitteringbehaviorresultsfromcontextualvariables,while85%resultsfromindividualvariability.Thisfindingshowsthatonanationallevel,thelargemajority(85%)oflitteringbehaviorresultsfromindividual‐levelvariables.Someexamplesofindividual‐levelvariablesincludelackofawareness,lackofconcern,orlackofmotivation,amongothers.Thisisnotsaythatphysicalcontextdoesnotmatter,andinfactourresultsshowthat15%ofthe


varianceinobservedlitteringbehaviorwasduetosomeaspectofthecontext(e.g.,existinglitter,lackofconvenientreceptacles,etc).

Oursecondsetofanalysesfocusedonindividual‐levelpredictorsoflitteringbehavior:age,gender,andwhethertheindividualwasinagroup.Ageshowedanegativelineartrend,withthehighestlitteringrate(26%)foundforyoungadults(age18‐29).Foradults30andolder,thelitteringrateremainedsteadyat~15%.Childrenandadolescents(youngerthan18)hadalitteringrateof13%.Whilegenderwasnotstatisticallysignificant,therewasatrendformen(21%)tolittermorethanwomen(15%).Giventhehistoricalinterestingenderasapredictoroflittering,weexploredthegendereffectinmoredepth.Whiletherewasnotanoveralleffectforgenderinthehierarchicallinearmodel,malesweremorelikelytolitterwithintentthanfemales(particularlyflick/fling).

Nootherindividual‐levelvariableswerepredictiveoflittering.However,the

variabilityintheLevel‐1equationremainedstatisticallysignificant,indicatingthatothervariablesarerequiredtofullyexplainindividualvariabilityinlittering.Weexplorethisissueinmoredetailusinginterceptsurveysandanationaltelephonesurvey(seebelow).

Usingthehierarchicalstructureofourdata,weproceededtoanalyzethecontextualpredictorsoflitteringbehavior.Weexamined10predictorvariables:

• temperature,• timeofday,• availabilityandnumberofreceptacles(trash,ashtray,orrecycling),• amountoflitterpresent,• physicalappearanceofthearea,• postedsignageaboutlitter,• numberofotherpeopleinthelocation,• weather,• locationtype(rural,urban,suburban),• andsitetype(e.g,citycenter,fastfood).

Atthesimplebivariatelevel,severalofthesevariablesweresignificant

predictorsoflitteringbehavior.However,whentheclusterednatureofthedatawastakenintoaccountstatistically,alongwiththeoverlapofthepredictors(forexample,physicalappearancewasgenerallycorrelatedwithloweramountsofexistinglitterr=‐.40andwithmoretrashreceptacles=.30),onlytwovariablesemergedasuniquelyandstatisticallysignificantpredictors:availabilityofdisposalreceptacles,andamountoflitterpresent.


Thefirstwastheavailabilityofandnumberofdisposalreceptacles.Aspartofoursiteobservations,theteamcountedthenumberofreceptacles(trash,recycling,cigarette,dumpster),alongwiththedistancefromthepersonatthetimeofdisposal.Theaveragewas5.8binsperlocation,witharangefrom0to19.Theanalysisforpresenceofreceptaclesrevealedanintuitivefindingthatwhennoreceptacleispresent,anydisposalisanimproperdisposal(e.g.,a100%litteringrate).Butmorerelevanttothecurrentresearchquestions,thestatisticalanalysisshowedthatlocationswithmorereceptacleshadalowerlitteringrate(γ=‐.01,p<.05).Thisstatisticalcoefficientcanbeinterpreteddirectly,suchthatforeveryaddedtrashreceptacle,thelitteringratedecreasedby1%(fromtheoverallrateof17%).

Thesecondstatisticallysignificantpredictoroflitteringbehaviorwasthepresenceoflitterinthesite.Locationswithmorelitterwereassociatedwithahigherlitteringrate.Thestatisticalanalysesshowedthatthepresenceofexistinglitter(bothcountedandratedbytheobserversonascalefrom0‐10)waspredictiveoflitteringbehavior(γ=.02,p<.001fortheobjectivecounts).

Noneoftheothercontextualpredictorswerestatisticallysignificant.However,

therewasatrendtowardgreaterlitteringinurban(23%),comparedtosuburban(18%),andrural(15%)locations.

Conclusions

Theresultsfromthestatisticalanalysesofthegenerallitterobservationssupportanumberofinterestingconclusions.Ofallthedisposalsthatweobservedat86locationsacrossthecountry,17%wereimproper.Thisisastrikinglyhighnumber,anddespitethestrongnormfavoringproperdisposalthathasemergedoverthepast40years(seephonesurveybelow),therateoflitteringremainsrelativelyhigh.Importantly,thislitteringrateisgeneratedfromarandomsampleofindividualsacrossarangeofdifferentlocations,andnotjustafewisolatedobservations.Whenanalyzedusingthetotalnumberofindividualsthatweobserved(8,990)wefindthat4%litter.Whilelow,thisisstillastrikingnumber—4%ofALLindividualsthatweobservedindiverselocationsacrossthecountry,littered.

Asecondimportantfindingpertainstothevariabilityinlitteringbehavior.Whilealargevolumeofdatahasbeencollectedaboutlitterandlitteringovertheyears,nonehasaffordedtheopportunitytosimultaneouslytestthedegreetowhichitisaffectedbypersonalandcontextualvariables.Ourfindingsindicatethat15%oflitteringactsresultfromcontextualvariables,and85%resultfrompersonalqualities.Thisfindingisparticularlyinstructive,becauseitindicatesthatgiventhesameinfrastructureandopportunitiestoproperlydispose,individualswillvarytremendously.Notethatifthetrendhadbeenreversed,suchthat85%ofthe


variancewasduetothesituation,itwouldindicatethatwhileindividualsvaryacrosssettings,withinasettingtheyactsimilarly(e.g.,litteringornot).

Finally,theresultsfromourstatisticalanalysesoflitteringbehavioridentifiedonlyacoupleofsignificantpredictors.Interestingly,genderwasnotasignificantpredictoroflitteringbehavior.Thisfindingrunscontrarytopriordatashowingthatmenaremorelikelytolitterthanwomen.Giventheobservationalnatureofthecurrentdata,wetendtofavortheexplanationthatmenaremorelikelytoreportlittering,butinrealitymenandwomenareequallylikelytodoit.Attheindividuallevel,wedidfindgoodevidencethatageisnegativelyrelatedtolittering,withyoungadultsmorelikelytolitterthanpersons30andover.

Atthelevelofthelocation,presenceoftrashreceptaclesandamountoflitter

presentweresignificantpredictorsoflitteringbehavior.It’stemptingtoaskaboutthe“optimal”numberofreceptaclesinalocation.Whileourdatadonotspeakdirectlytothisissue,wedohaveevidencethatthelowestlitteringrateoccurswhenreceptaclesareavailableandcloseathand.Indeed,thelitteringratewasonly12%ofdisposalsthatoccurredwithin10feetofareceptacle,comparedwitha30%litteringratefordisposalsmorethan60feet.

CigaretteLitterObservations

Inanefforttoexpanduponourgenerallitterobservations,theActionResearchteamconductedasmallersampleofobservationsfocusedonsmokers.Notethattheseobservationswereconductedinadditiontothesmokerscapturedaspartofthegenerallitterobservationsreportedinthepriorsection.AsshowninTable21(AppendixA),smokerobservationswereconductedat44locations,offoursitetypes:recreation(11sites),bar/restaurant(12sites),retail(12sites),andmedical/hospital(8sites).Theteamalsomadeoneunplannedsetofcigaretteobservationsatacitycenter.Aswithourgenerallitterobservations,thestudywasdesignedasahierarchicalmultilevelmodel,andobservationswereconductedusingadetailedprotocol.

Smokerswererandomlyselectedwithineachlocation,toensurearepresentativesample.Atotalof767observationsweremadeofsmokers.Onehundredandtwenty‐sevenobservationswereconductedinaruralarea(17%),230observationswereconductedinasuburbanarea(30%),and410observationswereconductedinanurbanarea(54%).Twohundredandfourteenobservations(28%)wereconductedinthemorning,before12:00p.m.,435observations(56%)wereconductedintheafternoonbetweenthehoursof12:00p.m.and3:59p.m.,112observations(15%)wereconductedintheeveningbetweenthehoursof4:00p.m.and7:59p.m.,and6observations(1%)wereconductedatnightafterthehourof8:00p.m.


Therewere412malesand344females,ranginginobservedagefrom21to72(M=40.64,SD=13.00;11notcoded).Whendisposing,thesamplewasfairlyevenlysplitbetweenbeinginagroup(N=367)oralone(N=391).Groupsrangedinsizefrom2to21,withthemostfrequentgroupsbeinggroupsoftwo(N=205),three(N=79),andfour(N=46).Ninewerenotcodable.

Ofthe767individualsthatwereobserved,206leftthesitestillsmokingand31werenotcodable(observerscouldnotdeterminewhathappenedtothecigarettebutt);187properlydisposedofthebutt;and343improperlydisposed.Thisresultedinalitteringrateof65%,slightlyhigherthanthe58%ratefoundinthegenerallitterobservations.

Themostfrequentmethodofdisposalwastoplacethecigarettebuttonthe

ground(N=303),followedbyleavingthesitewithanunfinishedcigarette(N=206),orplacingitinanashtray(N=119).Table9showsthefrequencyofdisposalmethodtypeusedforcigarettebuttlitter.

Table9.FrequenciesofDisposalMethodsforCigaretteButtLitter

Frequency PercentageGround† 303 40%LeftSitewithCigarette 206 27%Ashtray 119 15%TrashCan 31 4%Unknown/Unseen 31 4%Pocketed 18 2%Other† 16 2%RecyclingBin 16 2%Planter† 12 1%Bushes/Shrubbery† 7 1%On/AroundTrashReceptacle†

5 1%

FieldStrip 3 1%TOTAL 767 100%Note:†designatesdisposalscodedaslittering.

Whenitemswerelitteredontheground,dropwithintentwasthemostfrequentlyusedstrategy(N=122),followedbyflick/fling(N=94).Table10showsthefrequencyoflitterstrategiesused.


Table10.FrequenciesofLitterDisposalStrategiesforCigaretteButtLitter

Frequency PercentageDropwithIntent 122 35%Flick/Fling 94 27%Stomp 92 27%Other 17 5%Bury 9 3%Wedge 6 2%Shoot&Miss 3 1%TOTAL 343 100%

Atthestartofeachobservation,themeandistancetothenearesttrashreceptaclewas31feet(SD=29),andthemeandistancetoanashreceptaclewas30feet(SD=29).Atthetimeofdisposal,improperdisposerswereanaverageof31feetfromanashtray.

UnderstandingCigaretteButtLittering

Ourdataanalyticstrategyfollowedthemulti‐levelmodel,inwhichweexaminedboththeindividualandcontextualvariablesthatarepredictiveoflittering.TheanalysiswasconductedusingHierarchicalLinearModeling,inwhichtheindividualwastreatedasLevel1(gender,age,inagroup)andthesitecharacteristicswereanalyzedasLevel2(e.g.,presenceofashreceptacles,presenceoflitter,physicalappearance,rural/urban/suburbanlocation,signage).Theanalysiswasconductedusingthe530observedcigarettedisposals(187properdisposals,343improper).

Theinitial“null”modelshowedthattheoveralllitterratewas.66.Acrossthe530individuals(σ=.135,Z=15.70,p<.01)andthe44locations(τ00=.081,Z=3.65,p<.01),therewasconsiderablevariabilityinthelitteringrate.TheIntraClassCorrelationcoefficient(ICC)was.38.Thisstatisticisdirectlyinterpretableanditindicatesthat38%ofthevarianceincigarettelitteringbehaviorresultsfromcontextualvariables,while62%resultsfromindividualvariability.Examplesofindividual‐levelvariablesincludeawareness,concern,ormotivation,andexamplesofcontextualvariablesincludepresenceofreceptacles,convenience,orcleanlinessofthearea.Theobservedclusteringeffect(38%)isconsiderablyhigherthanthatobservedforgenerallitteringbehavior(15%),anditsuggeststhatmorestructural‐levelinterventionsareneededforsmokers(seeanalysesbelow,andrecommendations).


Followingthenullmodel,weproceededtotestforindividual‐levelandsite‐levelpredictorsofimproperdisposal.Attheleveloftheindividual,wetestedgender,age,andbeinginagroup.Theonlystatisticallysignificanteffectwasage,withyoungersmokersmorelikelytolitterthanolder(β=‐.01,p<.01).Forclarification,agewascodedintodemographiccategories.Thehighestlitteringratesoccurredforsmokersintheir20s(.66,N=150)and30s(.72,N=123),comparedtosmokersintheir40s(.58),50s(.66),or60s(.50).

Pursuanttothehierarchicalnatureofourdata,weproceededtoanalyzethecontextualpredictorsoflitteringbehavior.Weexaminedninepredictorvariables:

• physicalappearance,• presenceofsignageaboutlitter,• temperature,• timeofday,• availabilityofashreceptacles(totalnumberofashreceptaclesinthe

location),• amountoflitterpresent(anytype,notjustcigarette),• weather,• locationtype(rural,urban,suburban),• andsitetype(e.g,citycenter,fastfood).

Whilethesimplebivariaterelationshipsrevealedanumberofsignificant

correlations,themultilevelmodelidentifiedthreeuniquelypredictivevariables:location,existinglitter,andpresenceofashreceptacles.Oneofthestrongestpredictorsofcigarettelitteringwasthenumberofashreceptacles(γ=‐.09,p<.01).Theparameterestimatefromtheanalysisisdirectlyinterpretableandindicatesthatforeveryaddedashreceptacle,thelitteringrateforcigarettebuttsdecreasedby9%(fromtheinitialbaselitteringrateof65%).

Asecondpredictorofcigarettelitteringwastheamountofexistinglitter(γ=.06,

p=.03).Morelitteredenvironmentsattractedmorelittering.Thisfindingoccurredforbothratedamountoflitter(ona0‐10scale)andcountedlitter.Inaddition,theresultsshowedthatthetypeofexistinglitterwasnotrelatedtocigarettelitteringbehavior.Thatis,thepresenceofexistinglitterwaspredictiveoflitteringbehavior,butthetypewasirrelevant.Smokersweremorelikelytolitterifthecontextcontainedanytypeoflitter,notjustcigarettebutts.

Finally,thethirdpredictorofcigarettelitterwaslocationtype.Retaillocations

wereassociatedwiththelowestlitteringrate(.58),followedbycitycenters(.58).Barsandrestaurantswerethird(.62),whilerecreational(.74),andmedical/hospitalsites(.75)hadthehighestlitteringrates.Howeverit’simportanttopointoutthatwhilethedifferenceswerestatisticallysignificant,theoveralllitteringratewasquitehigh—morethan50%ofallsmokersthatweobservedlittered,regardlessof


location.

Conclusions

Theresultsforthesmokingobservationssupportseveralinterestingconclusions.First,theoveralllitteringrateforcigarettebuttswas65%.Thatis,across44locationsand767observedsmokers,65%ofthecigarettedisposalswereimproper.Consistentwiththishighrateofimproperdisposalwasthefindingthatcigarettebuttswerethemostfrequentlyobserveditemofexistinglitterwhenourresearchteamcanvassedthesitespriortoobservation(seeTable3).Andthatamongthe130siteswevisitedacrossthecountry,theaveragesitecontained69litteredbutts.

Asecondnotablefindingwasthehighdegreeofconsistencyofindividualsacrosssites.Whenwestatisticallyanalyzedthedata,wefoundthat38%ofcigarettebuttlitteringwasassociatedwiththecontext,while63%wasattributabletoindividualsources.Thissuggeststhatcigarettebuttdisposalwasmoreofaproductofthesituation,thanwasgenerallittering(wheretheICCwas15%).Thefindinghasimportantimplicationsforlitterpreventionstrategies,anditsuggeststhatbothstructuralandmotivationalapproachesareneeded.Atthestructurallevel,decreasingtheamountofexistinglitterandincreasingtheavailabilityofashreceptaclesemergedaskeycontextualpredictorsoflittering.Inaddition,individual‐levelmessagesandactivitiesthatpromoteapersonalobligationnottolitterareimportantfortargetingthemotivationalsideoftheequation.

Consistentwiththisfinding,ourstatisticalanalysesrevealedseveralmeaningfulpredictorsofcigarettebuttlittering.Aswiththegenerallitteringobservations,litteringwasnotrelatedtogender,andmenandwomenwereequallylikelytolittercigarettebutts.Andalsoconsistentwithourgenerallitterobservations,agewasnegativelyrelatedtolittering,suchthatyoungeradultsweremorelikelytolitterthanolderadults.Inadditiontotheseindividual‐levelpredictorsoflittering,wealsoexaminedthesite‐levelpredictors.Herewefoundtwoimportantpredictorsoflittering.Whileourinitialunivariatestatisticsrevealeddifferencesinlitteringratebylocations(e.g.,morelitteringatmedicalandrecreationalsites,andlessatretailandcitycenters),thesedifferenceswereentirelyexplainedbytwoaspectsofthesite:existenceandnumberofashreceptacles,andtheamountofexistinglitter.Notethatthesignificanteffecthereisforashreceptacleandnotanyreceptacle.Thatis,thepresenceoftrashcansorrecyclingcontainersdidnotreducetherateofcigarettebuttlittering.Itwasonlythepresenceofashreceptacles,eitherasstand‐alone,orintegratedintoatrashcan,thatcorrelatedwithlowerratesofcigarettebuttlittering.


PARTIII:INTERCEPTINTERVIEWS

INTRODUCTION

Inthesecondpartofourresearchproject,weconductedinterceptsurveysatrandomlyselectedobservationsites.Thepurposeoftheinterceptsurveyswastoenhancetheresultsfromtheobservations,allowingustobetterunderstandtheattitudesaswellasthebehaviorsofindividualswholitter.Ourgoalhereistoreportboththebasicfrequencyandpatternofresponsesfromtheintercepts,andtoexaminetherelationshipbetweenthesevariablesandobservedlittering.

METHODS

SelectionofInterceptSites

Thelocationoftheinterceptsiteswasdeterminedatrandomandincludedbothgenerallitterandsmokingsites.Ourgoalwastoobtainasmall,representativesampleofindividualsandtolinktheresponsestoourinterceptinterviewtotheobserveddisposalbehaviors.

InterceptProtocol

Ateachoftheselectedinterceptlocations,individualswereapproachedtotakepartinaface‐to‐facesurvey.Fortheinterceptsites,thefieldresearchteammembersweredefinedaseitherobserversorinterviewers.Observerswatchedindividualswithinaspecifiedboundaryusingtheprotocolsummarizedearlierinthisdocument.Followingtheobservation,theobservercommunicatedwiththeinterviewer,viaaBluetoothhands‐freedevice,anddescribedthepersonselectedfortheinterview.Theinterviewerthenapproachedtheselectedpersontoaskforparticipationinashortoralsurvey(lessthanfourminutes)andindividualswereoffered$5inreturnfortheirparticipation.Theinterviewerdidnotknowifthepersonofinterestlitteredordidnotlitter.Theintervieweraskedthepersonasetofpre‐determinedquestions,includingdemographicitemsandquestionsaboutlitterbehavior,adopteddirectlyfromthenationaltelephonesurvey(reportedbelow).SeeAppendixCforthefullinterceptsurveyusedinthisstudy.


RESULTS

Ourfocushereisonthe102interviewsconductedin15generallittersites.Acrossthe15sites,196individualswererandomlyidentifiedforaninterview.Ofthe196,102completedthesurvey,58leftthesitebeforecontactcouldbemade,and36refusedtocompletetheinterview(resultingina52%responserate,and74%completionrate).Nineteenpercentoftheinterceptswereconductedinaruralarea,46%wereconductedinasuburbanarea,and36%wereconductedinanurbanarea.Fifty‐fivepercentofparticipantsweremaleand45%werefemales.Agesoftheparticipantsrangedfrom19to71(M=36,SD=13).

Whenaskedhowimportanttheissueoflitterwasinthegivenlocation,onascaleof0(notatallimportant)to10(extremelyimportant),theaveragerespondentsaidthatitwasveryimportant(M=8.27,SD=2.04).Whenaskedwhatpercentageofpeoplelitteredinthislocation,theresponsesrangedfrom0to90%,withanaverageresponseof40%(SD=27.66).

Inthepastyear,40%ofrespondentsreportedthattheyhadseenorheardsomethingaboutlitterprevention,41%hadseenorheardsomethingaboutscheduledcommunityclean‐up,25%hadseenorheardsomethingdiscouragingcigarettebuttlitter,and65%hadseenorheardamessagesimplytellingpeople‘donotlitter.’Forthoserespondentswhohadseenorheardanymessage,themostfrequentlyreportedplacewasonapublicserviceannouncement,commercial,ornewsreportonthetelevision.Table11showsthefrequencyofplaceswherethesemessageswereseenorheard.

Table11.ReportedLocationofMessaging

FrequencyPSAontheradio 5PSA/Commercial/NewReportonTV 21Billboard 12Newspaper 1CommunityFlyer/Bulletin 1

Whenaskediftheyhadlitteredinthepastmonth,57%ofrespondentsreportedthattheyhadnotlittered.Ofthe43%ofrespondentswhoreportedthattheydidlitter,themostfrequentlyreportedlittereditemwascigarettebutts(19%).Table12showstheitemsreportedtobelitteredbytherespondents.Themostfrequentlyreportedplacetolitterwasonthestreetorinaparkinglot.Interestingly,onlythreerespondentsexplicitlyreportedthattheyhadlitteredatthecurrentlocation.Table13showsthesereportedplacesoflittering.


Table12.ItemsReportedLittered

FrequencyBeverageBottle:Plastic 1BeverageBottle:Glass 1BeverageCan 3FoodWrapper 8FoodRemnant 4CigaretteButt 21Paper 5MixedTrash 1Other 1

Table13.RespondentReportedLitteringLocations

FrequencyEverywhere 1Vehicle 1Beach 1BusStop 1Other 3

Street/ParkingLot 21N/A 2Ground 2Home 2Park 3School 1Stairwell 1

Fifty‐fivepercentofrespondentsreportedthattheysmokedcigarettes.And,77%ofrespondentsreportedthattheythoughtcigarettebuttswerelitter.Whenaskedhowtheydisposedoftheircigarettebuttswhentheywereoutside,anashtraywasthemostcommonlyreportedplaceofdisposal.


Table14.PlaceofDisposalforCigaretteButtLitter

FrequencyPocketed 2Trashcan 8Ashtray 26Ground 7FieldStrip 5PersonalAshtray 2CigarettePack 1

Finally,personalanti‐litteringobligationwasexamined.Therespondentswereaskedaseriesoffivequestionsregardingtheextenttowhichtheyfeltapersonalobligationtonotlitter.Thesefivequestionswereaveragedintoacompositescore,rangingfrom0(nopersonalobligation)to10(veryhighpersonalobligation).Mostrespondentsreportedhavingahighpersonalobligationtonotlitter(M=8.61,SD=1.90).

LinkingInterceptsandObservations

Ofthe102respondentswhocompletedthesurvey,23wereobservedtohavelittered(23%).Ofthese23observedlitterers,15reportedlitteringinthepastmonthwhile8deniedlittering.Thatis,35%oftheindividualswespokewithdeniedlitteringinthepastmonth,wheninfact,wehadjustseenthemdoso.ThisfigureisnearlyidenticaltothefindingsfromasimilarstudyinAustralia,where37%ofindividualsobservedlitteringdeniedlitteringinthepast24hours.Thefindingsuggeststhatrelyingonselfreportoflitteringbehaviorisproblematic,andthattherewillbeageneraltendencytounderreportthebehavior.

Inanefforttoexpandonourearlieranalysesofobservedlittering,weconductedaregressionanalysisusingtheinterviewitemstopredictobservedlittering.Inthisanalysis,weusedfiveindividual‐levelpredictors(age,gender,personalobligationtonotlitter,attitudesaboutlitter,andperceptionsaboutthefrequencythatotherpeoplelitterinthislocation).Fromthisanalysis,twovariablesemergedasimportantpredictorsofobservedlittering:age(beta=‐.26,p=.01),andpersonalobligationtonotlitter(beta=‐.22,p=.02).

Similarresultswerefoundfromananalysisusingself‐reportedlitteringbehavior.Agewasanevenstrongerpredictorofself‐reportedlittering(beta=‐.40,p<.001),withyoungeradultsmuchmorelikelytoreportlitteringthanolderadults.Inaddition,apersonalobligationtonotlitterwasalsosignificantlypredictive(beta=‐.24,p<.01).


PARTIV:TELEPHONESURVEY

INTRODUCTION

Althoughagreatdealofresearchhasbeendoneonlittering,veryfewstudieshavelookedatself‐reportedlitteringhabitsonanationallevel.ThistelephonesurveywasdevelopedincooperationwithKeepAmericaBeautifulandisthefirstnational‐levellittersurveyin40years.

Inthe1968study,titled“WhoLitters–andWhy,”theresultsfromtwonationalsurveysarereported.Whilethedataanddetailedresultsarelosttohistory,therewereseveralbroadfindingsthatwarrantcomment.Theresultsshowedthatdespitehighreportedratesoflittering,alargemajorityofthepublicsawlitteringasaproblem,andsupportedlawsandfinestoreducelittering.Approximately50%ofthesampleadmittedtooneormorelitteringpractices,andlitteringwasmostcommonamongyoungmales.Thusthisearlyresearchshowedamarkedgapbetween“awareness”andbehavior,andthereportconcludedthat:

“thepublicbelievesthereshouldbelawsagainstthisevil,andamajoritybelievethatthereshouldbegreaterenforcementofthelaws.Butthesamepersonwhodescribesalittererasa“slob”andbelievesheshouldbearrested,still—inalmostthenextbreath—saytheythemselvesareguiltyoflittering.”

Thegoalofthesurveywastobuildonthisearlystudy,andtoprovidearesearchfoundationforthevariablesaffectinglitteringbehavior.Itemswerecreatedtomeasurelitteringbehaviorinconjunctionwithvariousenvironmental,demographic,andpsychologicalfactors.Athoroughreviewofcommunityandstate‐levelsurveyswasconductedinordertoincludecommonlyusedmeasures,wherepossible.Inaddition,wecreateduniquemeasuresforcommunitylivabilityandlitterbehaviorbytypeandlocation.

TheinstrumentisintendedforusebyKABaffiliatesandcommunityleaderswhenregionaldataareneededformonitoringorcomparison.Theresultsalsoprovidebaselinedataforcurrentprograms,aswellasfoundationsforthedevelopmentofnewprograms.ThetelephonesurveyitemscanbefoundinAppendixD.


METHODS

Wecollected1,039telephoneinterviewsbetweenMay29andJune16,2008.Inanefforttoobtainarepresentativesample,weemployedarandom‐digit‐dialingsamplingmethod.Telephonenumbersforallhouseholdsnationwidehadanon‐zerochanceofbeingselected.Usingthismethod,listedandunlistednumberswereincludedinthesample.Theonlyqualificationforparticipationwasthattherespondentbeeighteenyearsofageorolder.

Data

Thesamplewasmadeupof360men(35%)and679women(65%).Respondentagerangedfrom18to101,withameanageof54years.SevenpercentdescribedthemselvesasHispanic,8%asBlackorAfricanAmerican,and81%aswhite.Overathirdoftherespondentsreportedbeingcollegegraduates(38%),while56%werehighschoolgraduatesand6%reportednotgraduatingfromhighschool.Overhalfoftherespondents(55%)statedtheycurrentlyworkforpay.Eighteenpercent(18%)oftherespondentswereclassifiedascurrentsmokers.

Measures

Themainobjectiveoftheanalysiswastocreateameasureofself‐reportedlitteringbehaviorinordertoidentifythevariablesthatarerelatedto,aswellaspredictlitteringbehavior.Aseriesofanalysesofvariance(ANOVAs)wereusedtolookatlitteringbehavioracrossvariousgroupedvariables.Thenextstepwastoconductamultipleregressionanalysistolookatthecontributionofseveralpredictorvariablesonself‐reportedlittering.Scaleswerecreatedinordertomeasureself‐reportedlitteringbehavior,communitylivability,andpersonalnormsagainstlittering.

KABLitteringItems.EightquestionsfromtheKAB1968telephonesurveywereincludedinthisstudy.Respondentswereaskedifoverthepastmonththeyhaddropped,leftbehind,orthrownvariousmaterials(gumorcandywrappers,facialtissue,foodorfoodwrappers,beerorsodacansorbottles,papercontainers,newspaper,orpaperorfoodremnants).

SelfReportedLitteringBehavior.Themajorityoftheanalysesusealitteringbehaviorscalecreatedtounderstandthevariablesthataffectlitteringbehavior.Respondentswereaskedtoratethelikelihoodtheywouldlitterfivedifferenttypesoflitter(applecoreorbananapeel,gumorcandywrapper,foodcontainer,gum,andcigarettebutt),acrossthreedifferentlocationtypes(outavehiclewindow,onthegroundwhilewalkingtoavehicleortransitarea,andonthegroundwhenatapark


oroutdoorarea).Allrespondentswereaskedthetwelvemainquestions(fourlittertypesbythreelocationtypes)andsmokersreceivedanadditionalthreequestionsregardingcigarettebuttdisposal.Theresponsesrangedfromzerototen,wherezeroequalednotatalllikelytolitterandtenequaledverylikelytolitter.Theoveralllitteringbehaviorscalewasconstructedbycalculatingthemeanoftheratingscores.Thescoresrangefromzerototen,andhigherscoresrepresentmorelitteringbehavior.

Additionally,separatescaleswereconstructedtorepresenteachlittertypeandeachlocationtype.Forexample,thethreequestionsthataskedaboutthelikelihoodofdroppingfoodcontainers(acrossthethreelocationtypes)wereaveragedtocreateafoodcontainerscale.Thisgaveusinformationaboutlitteringbehavioracrosslitterandlocationstypes.Foreachscale,therangeofscoreswasfromzerototen.

CommunityLivability.Communitylivabilityquestionswereconstructedtoexaminetheeffectsofclean,safe,andattractivecommunitycharacteristcsonlitteringbehavior.Respondentswereaskedeightquestionswhichmeasuredvariousdimensionsofthecurrentstateoftheircommunity.Althoughnostandardscaleexiststomeasurecommunitylivability,extensiveresearchwasdonetoensurethescaleincludeditemsthatwouldreflectthefullrangeofissues.Theitemsmeasuredcommunitycleanliness,structuralmaintenance,walkability,safety,landscaping,andinfrastructure.Eachquestionconsistedoffourresponsecategoriesandtheratingsfromeachoftheeightquestionswereaveragedtoproduceascore—alivabilityquotient.Therangeofscoresforthescalewasfromonetofour,withhigherscoresindicatingratingsofgreaterlivability.

PersonalNorms.RespondentswereaskedanabbreviatedversionofthePersonalNormsAgainstLitteringScale(Kallgren,Reno,andCialdini,2000).Thisscalewasusedtomeasuretheeffectofpersonalnormsonlitteringbehavior.Asetoffivesituationalquestionswereaskedtomeasurepersonalobligationnottolitter.Ratingsontheindividualquestionsrangedfromzero(nopersonalobligation)toten(averystrongpersonalobligation).Theratingswereaveragedtoprovideeachrespondentwithapersonalobligationagainstlitteringscore.Scoresrangedfrom0to10,withhigherscoresindicatingastrongerpersonalobligationnottolitter.

WorkEnvironment.Inordertounderstandtherelationshipbetweenworkenvironmentsandlitteringbehavior,respondentswhostatedtheycurrentlyworkedforpaywereaskediftheyworkedinanindoororoutdoorenvironmentandwhethertheareawasdesignatedsmoke‐freeornot.

CurrentSmokers.PrevalencedatafromtheCentersforDiseaseControlandPrevention(CDC)showsthatin2006,20percentofAmericanadultsweredefinedascigarettesmokers.Ofparticularinterestinthisresearchwascigarettebuttlitteringhabitsofsmokers.WeusedtheCDCquestionstoidentifyrespondentsascurrentsmokers.Respondentswereaskediftheyhadsmokedatleast100


cigarettesintheirentirelife,andifso,iftheycurrentlysmokedalldays,somedays,ornotatall.Respondentswhoindicatedtheysmokedalldaysorsomedaysweredefinedasacurrentsmoker.Currentsmokerswerethenaskedasetofquestionsaboutcigarettebuttdisposal.Theywereaskedwhethertheyhadreceptaclesatworkorintheirvehicle,andwhetherornottheyownedapocketashtray.

LitterMotivatorsandBarriers.Allrespondentsweregivenasetofsixitemsregardinglittermotivatorsandbarriers.Thestatementswereanattempttounderstandsomeofthebarriersandmotivatorstoproperlitterdisposal.Respondentswereaskedtoindicatethedegreetowhichtheyagreedordisagreedwithstatementssuchas,IammorelikelytolitterwhentheitemI’mholdingisn’trecyclable,andIammorelikelytolitterwhenIaminabadmood.Theresponsecategorieswere:stronglyagree,agree,disagree,andstronglydisagree.

LitterImportanceandMessages.Lastly,respondentswereaskedabouttheimportanceoftheissueoflitterintheirtownorcity,andwhetherornottheyhadseenorheardanylitterpreventionmessagesinthepastyear.Forthosewhostatedtheyhadseenorheardlitterpreventionmessages,theywerethenaskedthesubject‐matter(e.g.,communityclean‐uporprevention)andwheretheysaworheardthemessages(e.g.,television,radio,billboard,orposter).


RESULTS

LitterBehavior

Thesurveycontainedtwotypesofself‐reportedlitterbehaviorquestions.Thefirstsetconsistedofeightdichotomous(yes–no)itemsadoptedfromthe1968KABsurveyoflitteringinAmerica,whichaskedifthebehaviorhadoccurredinthepastmonth.Table15showstheresults.It’simportanttonotethatthefrequenciesandpercentagesreportedinTable15arefortheentiresample,andnotjustindividualswhoreportedlittering.Thus,thepercentagesshownintheTablereflectagenerallylowlevelofreportedlitteringbehavior.

Table15.SelfReportedLitteringBehaviorinthePastMonth

Frequency PercentageDroppedGumWrappersonGround 68 6.58%LeftNewspaperBehind 53 5.11%ThrownThingsOutofCarorBoat 42 4.05%DroppedFoodorWrappersonStreet 38 3.67%DroppedPaperContainersonGround 25 2.41%DroppedFacialTissueonGround 14 1.35%LeftPaperorFoodatPicnicArea 16 1.55%DroppedCansorBottlesOutdoors 9 0.87%

N=1,039

Eighty‐fivepercentoftherespondentsreportednoneofthelistedlitteringbehaviorsinthepastmonth(N=876).Twelvepercentreportedparticipatinginoneortwobehaviors(N=127),while2.5%reportedparticipatinginthreeormorebehaviors(N=25).Theresultingdatashowedthatthereportedlitteringratein2008was15%,comparedwitha50%litteringratein1968.

Theprimaryfocusofouranalyseswasonthesecondmeasureofself‐reportedlitteringbehavior.Respondentswereaskedthelikelihoodtheywoulddropspecificitems(applecoreorbananapeel,gumorcandywrapper,foodcontainer,gum,andcigarettebutt),acrossspecificlocations(outavehiclewindow,onthegroundwhilewalkingtoavehicleortransitarea,andonthegroundwhenataparkoroutdoorarea).Thelitteringbehaviorscaleprovidesusacombinedmeasureofallthebehaviors,aswellasmeasuresofbehaviorforthespecifictypesoflitterandatthespecificlocations.Figure3showsthemeanscoresfortheoveralllitterbehaviorscale,thelittertypescales,andthelocationtypescales.Thescalerangeisfromzerototen.


Figure3:MeanLitterScoresforOverallLitterBehavior,LitterType,andLitterLocation

Themeanscorefortheoverallbehaviorwas.61(SD=1.45),ona10‐pointscale.Respondentsreportedthegreatestlikelihoodoflitteringbehaviorwithcigarettebutts(M=1.63,SD=2.87),applecoresandbananapeels(M=.86,SD=1.98).Respondentsweretwiceaslikelytoreportdroppingacigarettebuttasanapplecoreorbananapeel.

LitterBehaviorandCommunityLivability

Analysisofvariancewasperformedtoidentifygroupdifferencesinlitteringbehaviorwithineachoftheeightcommunitylivabilityvariables.Table16showsfrequenciesfortheresponses,alongwithmeansandstandarddeviationsforeachofthecommunitylivabilityitems.


Table16.GroupMeansofLitteringBehaviorbyCommunityLivability

Item FrequencyMean

Littering SDVisibleLitter(QBI1)**

Virtuallynovisiblelitter 355 0.53 1.47Uponcarefulinspectionasmallamountoflitterisobvious 526 0.60 1.33Visiblelittercanbeseenthroughoutthearea 130 0.67 1.49Majorillegaldumpsitesarepresent 21 1.77 2.84TOTAL 1032 0.61 1.45HowCleanisCommunity(QBI2)*

Veryclean 553 0.59 1.58Somewhatclean 424 0.58 1.20Notveryclean 43 0.81 1.20Notatallclean 8 2.15 3.67TOTAL 1028 0.61 1.45StreetsandSidewalksMaintained(QBI3)

Verywellmaintained 509 0.57 1.48Somewhatmaintained 308 0.69 1.42Notverywellmaintained 87 0.67 1.44Notatallmaintained 44 0.48 0.85TOTAL 948 0.62 1.43StreetsandSidewalksSafeDuringDay(QBI4)**

Verysafe 665 0.61 1.47Somewhatsafe 238 0.49 1.08Somewhatunsafe 46 0.95 1.83Veryunsafe 20 1.74 2.78TOTAL 969 0.62 1.46StreetsandSidewalksInviting(QBI5)

Veryinviting 455 0.60 1.56Somewhatinviting 370 0.59 1.22Somewhatuninviting 58 0.92 1.84Veryuninviting 54 0.75 1.23TOTAL 937 0.63 1.44NumberofFlowers,Plants,andTrees(QBI6)

Therearealotofflowers,plants,andtrees 551 0.57 1.42Therearesomeflowers,plants,andtrees 362 0.64 1.47Therearenotmanyflowers,plants,andtrees 85 0.68 1.43Therearenoflowers,plants,andtrees 29 0.75 1.98TOTAL 1027 0.61 1.45AttractivenessofFlowers,Plants,andTrees(QBI7)*

Theyareveryattractive 547 0.62 1.58Theyaresomewhatattractive 387 0.54 1.14Theyaresomewhatunattractive 57 0.67 1.04Theyareveryunattractive 26 1.44 2.76TOTAL 1017 0.61 1.45


AttractivenessofInfrastructure(QBI8)*

Theyareveryattractive 316 0.60 1.49Theyaresomewhatattractive 486 0.58 1.38Theyaresomewhatunattractive 116 0.58 1.12Theyareveryunattractive 41 1.31 2.35TOTAL 959 0.62 1.45Note:Frequency=numberofoccurrences,Mean=average,SD=standarddeviation.*differencessignificantlyatthep<.05level,**differencesignificantatthep<.01level

Significantdifferenceswerefoundforgroupsinallvariables,exceptforthestreetsandsidewalksarewellmaintainedforpeoplewhoarewalking(QBI3),thestreetsandsidewalksinyourcommunityareaninvitingplaceforpeopletowalk(QBI5),andthenumberofflowers,plants,andtreesinyourcommunity(QBI6).Participantswerethreetimesaslikelytoreportlitteringbehavioriftheylivedincommunitieswithamajoramountoflitter(QBI1),thaniftheylivedincommunitieswithvirtuallynovisiblelitter.Thosewhoreportedlivingincommunitiesthatwereverycleanandsomewhatclean(QBI2)weremuchlesslikelytoreportlitteringbehaviorthanthosewholivedincommunitiesthatweresomewhatuncleanandveryunclean.Thesafetyofstreetsandsidewalksduringdaylighthours(QBI4)alsoshowedsignificantdifferencesbetweengroups.Thosewhoreportedlivinginsomewhatunsafeandveryunsafecommunitiesreportedsignificantlymorelitteringbehavior.Whererespondentsreportedlivingincommunitieswhereflowers,plants,andtrees(QBI7)wereveryunattractive,litteringbehaviorwassignificantlygreaterthanintheotherthreegroups.Thesamepatternwastruefortheattractivenessoftheinfrastructure,thingslikebenches,planters,streetsigns,andstreetlights(QBI8).Respondentswhoreportedlivingincommunitieswheretheinfrastructurewasveryunattractive,reportedsignificantlyhigherratesoflitteringthanfortheotherthreegroups.

Conclusion

Anoverallpatternemergedbetweenseveraldimensionsofcommunitylivabilityandlitteringbehavior.Thesedatasuggestthatthoselivingincommunitieswherethereislessvisiblelitter,thecommunityisclean,thestreetsandsidewalksaresafetowalkduringdaylighthours,andtheplants,aswellastheinfrastructureareattractive,arelesslikelytoreportlitteringbehavior.Althoughthishasbeensuggestedinotherliterature,astandardinstrumentformeasuringcommunitylivabilitydoesnotyetexist.Theresultsreportedaboveunderscoretherelationshipbetweencommunityappearanceandlitteringbehavior.


LitterBehaviorandCigaretteSmokers

Currentsmokerscomprised18.4%ofthesample(N=191).Thosecurrentsmokerswhoworkedforpay(N=170)wereaskediftheirworklocationhadreceptaclesforcigarettebutts(QCD1);41.8%statedtheirworklocationdidnothavereceptaclesforcigarettebutts.Currentsmokerswerealsoaskediftheyhadareceptacleforcigarettebuttsintheircar(QCD2);28.1%indicatedtheydidnothaveareceptacleintheircar.Fourteenpercentofthecurrentsmokersreportedowningapocketashtray(QCD3)(N=27).

Analysisofvariancewasusedtounderstandthelitteringbehaviordifferencesbetweennon‐smokersandcurrentsmokers.Figure4displaysthemeandifferencesforlitteringbehaviorbetweennon‐smokersandcurrentsmokersusingthelittermeasureincludingcigarettebuttdisposal,excludingcigarettebuttdisposal,andusingageasacovariate.Thescalerangeisfromzerototen.

Figure4:GroupMeansforLitteringBehavior(010)byCurrentSmokerStatus

Usingthelitterbehaviormeasureincludingcigarettebuttdisposal,currentsmokers(M=.97,SD=1.90)reportednearlytwicethelitteringbehaviorasnon‐smokers(M=.53,SD=1.31).Thedifferenceisstatisticallysignificantatthep<.01level.Removingcigarettebuttdisposalfromthelitteringbehaviormeasurestillshowedsignificantdifferencesbetweencurrentsmokers(M=.80,SD=1.88)andnon‐smokers(M=.52,SD=1.31).Thedifferenceisstatisticallysignificantatthep<.01level.However,inordertofurtherexplaintherelationshipbetweensmoking


statusandlitteringbehavior,weperformedananalysisofvarianceusingageasacovariate.Theresultsshowthatageexplainsasignificantamountofthevariationinsmokingstatusandlitteringbehavior.Afteragewasintroducedasacoviariatewithsmokerstatus,themeandifferencebetweencurrentsmokersandnon‐smokerswasestimatedat.71and.54,respectively.

Conclusion

Atfirstglance,theresultsappeartoshowapredictiverelationshipbetweencurrentsmokerstatusandself‐reportedlitteringbehavior.However,becausetherelationshipsbetweensmokingandage,andlitteringbehaviorandagearenegative(asageincreases,bothsmokingbehaviorandlitteringbehaviordecreases),furtheranalyseswereconductedinwhichwestatisticallycontrolledfortheageeffects.Thesenewanalysesshowedthattherelationshipbetweensmokingandlitteringbehaviorwasconfoundedbyage.Whileatfirstglancethedataappeartoshowthatsmokingpredictslitteringbehavior,whenageisusedasacovariate,therelationshipbetweencurrentsmokingstatusandlitteringbehaviorbecomesnonsignificant.Thatis,thehighrateoflitteringamongsmokersisdueinparttothetendencyforsmokerstobeyoungerthannonsmokers.

LitterBehaviorandtheWorkEnvironment

Analysisofvariancewasusedtolookatgroupdifferencesacrosstwoworkplacevariables.Forthosewhoreportedworkingforpay,weexaminedlitteringbehaviordifferencesbetweenindoorandoutdoorworkenvironments(QWE2).Figure5showsthegroupdifferences.Thescalerangeisfromzerototen.

Figure5:GroupMeansforLitteringBehavior(010)byIndoorandOutdoorWorkEnvironment


Participantsworkinginanoutdoorenvironmentreportednearlytwicethelitteringbehaviorasthoseworkinginindoorenvironments.Thedifferencewassignificantatthep<.01level.

Therelationshipbetweenindoorandoutdoorenvironmentsandsmoke‐freeenvironments(QWE3)wasexaminedaswell.Figure6showsthelitteringbehaviordifferencesforthesefourgroups.Thescalerangeisfromzerototen.

Figure6:GroupMeansforLitteringBehavior(010)byIndoorandOutdoorWorkEnvironmentsand

SmokeFreeWorkplaces

Whilethoseintheoutdoorworkenvironmentgroupshowedthelargestmeanforlitteringbehavior,thesmoke‐freeworkplacegroups(indoorandoutdoor),reportlesslitteringbehaviorthanthoseworkinginenvironmentsthatarenotsmoke‐free.Again,thelitteringbehaviormeasurementincludescigarettebuttdisposal,butitisimportanttonotethattheseresultsarenotexclusiveofcigarettebuttdisposal.Thedifferencesaresignificantatthep<.01level.

Figure7showsthegroupmeansforcigarettebuttdisposalbehaviorinthesmoke‐freeworkplacegroupsandintheoutdoorworkenvironment.


Figure7:GroupMeansforCigaretteButtDisposalBehavior(010)bySmokeFreeandNotSmokeFreeWorkplacesand

OutdoorWorkEnvironment

Therewerenodifferencesforcigarettebuttdisposalbehaviorbetweenthoseworkinginasmoke‐freeworkplaceandthosenotworkinginasmoke‐freeworkplace.Whilethoseworkinginanoutdoorworkenvironmentshowedagreatermeanforcigarettebuttdisposalbehavior(M=3.0,SD=4.34),thedifferencewasnotstatisticallysignificant.

Conclusion

Thesedatashowthattheoutdoorworkenvironmentgroupreportedmorelitteringbehaviorthantheindoorworkenvironmentgroup.Respondentsnotworkinginsmoke‐freeworkplaceenvironmentsalsoreportedmorelitteringbehavior.Forcigarettebuttlitterdisposal,althoughworkingoutdoorsincreasedlitterbehavior,workinginsmoke‐freeworkenvironmentoroutdoorsdidnotsignificantlyalterbehavior.Withtheincreaseofsmoke‐freeworkenvironmentsnationwide,adequateinfrastructureforproperdisposalisanemergingissue.Thisisanimportanttopicforfutureresearch.

LitterBehaviorandtheMotivatorsandBarriers

Analysisofvariancewasusedtolookattherelationshipbetweenlitteringbehaviorandsixsituationalscenarios.Respondentswereaskediftheystronglyagreed,agreed,disagreed,orstronglydisagreedwiththelikelihoodtheywouldlitterinthefollowingsituations:(QLB1)whentheitemI’mholdingisn’trecyclable,(QLB2)whenIaminabadmood,(QLB3)whenIknowsomeoneelsewillbearoundtopickitup,(QLB4)whenIdon’thavetimetotakecareofitanyotherway,(QLB5)whenthereisn’tatrashcannearby,and(QLB6)whentheitemisbiodegradable.Table17displaysthemeansandstandarddeviationsfortheseitems.


Table17.GroupMeansforLitteringBehavior(010)byMotivatorsandBarriersItems

I’mmorelikelytolitter… FrequencyMean

LitteringStandardDeviation

WhentheitemI'mHoldingIsn'tRecyclable(QLB1)**

StronglyAgree 25 1.48 2.73Agree 61 1.77 2.29Disagree 350 0.71 1.75StronglyDisagree 585 0.39 0.89TOTAL 1021 0.61 1.45WhenI'minaBadMood(QLB2)**

StronglyAgree 16 3.28 3.57Agree 74 1.56 1.94Disagree 397 0.60 1.64StronglyDisagree 538 0.41 0.88TOTAL 1025 0.61 1.45WhensomeoneElseWillPickItUp(QLB3)**

StronglyAgree 16 2.01 3.44Agree 42 2.07 2.35Disagree 407 0.71 1.69StronglyDisagree 566 0.39 0.89TOTAL 1031 0.61 1.45WhenIDon'tHaveTime(QLB4)**

StronglyAgree 17 1.98 2.94Agree 89 1.50 1.80Disagree 410 0.70 1.78StronglyDisagree 506 0.33 0.77TOTAL 1022 0.61 1.45WhenthereIsn'taTrashCanNearby(QLB5)**

StronglyAgree 16 1.97 3.01Agree 74 1.84 2.16Disagree 440 0.66 1.66StronglyDisagree 494 0.34 0.78TOTAL 1024 0.61 1.45WhentheItemisBiodegradable(QLB6)**

StronglyAgree 44 1.57 2.09Agree 186 1.26 1.75Disagree 393 0.56 1.61StronglyDisagree 393 0.25 0.73TOTAL 1016 0.61 1.45

Note:Frequency=numberofoccurrences,Mean=average,SD=standarddeviation.**differencesignificantatthep<.01level


Significantlitteringbehaviordifferenceswerefoundacrossgroupsforallthemotivatorsandbarriersitems.Respondentswhoreportedtheyaremorelikelytolitterwhentheitemisn’trecyclable(QLB1)weretwiceaslikelytoreportlitteringbehaviorthanthosewhodisagreedwiththestatement.Thosewhoreportedtheyaremorelikelytolitterwhentheyareinabadmood(QLB2)reportedsignificantlyhigherratesoflitteringbehaviorthanthosewhodisagreedwiththestatement.Respondentswhostatedtheyaremorelikelytolitterwhensomeoneelsewillpickitup(QLB3)reportedmuchgreaterlitteringbehaviorthanthosewhoreportedtheywereunlikelytolitterinthatsituation.Respondentswhostatedtheywerelikelytolitterwhentheydidn’thavetimetotakecareofitanyotherway(QLB4)reportedsignificantlyhigherratesoflitteringbehaviorthanthosewhostatedthattimewasunlikelytobeafactor.Thosewhoreportedtheyweremorelikelytolitterintheabsenceofatrashcanorbag(QLB5)showedsignificantlygreaterratesoflitterbehaviorthanthosewhounlikelytolitterinthatsituation.Andlastly,respondentswhostatedtheyweremorelikelytolitterwhentheitemwasbiodegradable(QLB6)reportedmuchmorelitteringbehaviorthanthoseforwhombiodegradabilitywasunlikelytobeafactor.

Conclusion

Theanalysisindicatestherearepersonal(badmood,notime),situational(someoneelsewillpickitup,notrashcannearby),anditem‐specific(itemisn’trecyclable,itemisbiodegradable)evaluationsthataffectlitteringbehavior.Alldifferenceswerestatisticallysignificant.

LitterBehaviorandtheMessages

Analysisofvariancewasusedtoexploretherelationshipbetweenlitteringbehaviorandlitterpreventionmessages.Figure8showsthegrouplitteringbehaviordifferencesforthosewhoreportedtheyhadandthosewhoreportedtheyhadnotseenorheardlitterpreventionmessagesinthepastyear.Thescalerangeisfromzerototen.


Figure8:GroupMeansforLitteringBehavior(010)bySaworHeardLitterPreventionMessages

Therewasnosignificantdifferencebetweenthosewhoreportedtheyhadseenorheardlitterpreventionmessagesinthepastyearandthosewhoreportedtheyhadnot.Whilethisfindingissomewhatdiscouraging,wedonotinterpretitasevidencethatmessagingisineffectiveatreducinglitteringbehavior.First,theoverallreportedrateoflitteringisextremelysmall(.5onascalefrom0to10).Second,thefindingsfromourobservationssuggestthatself‐reportedlitteringratesaresubjecttoreportingbias,andmightnotaccuratelyreflectactuallitteringrates.Indeed,thesameindividualswhoarelikelytounderreportlitteringarelikelytoover‐reportexposuretolitterpreventionmessages.Andthird,wedonotdifferentiateherebetweendifferenttypesoflitterpreventionmessages,andsomemessagesaresurelymorepowerfulataffectbehaviorthanothers(seerecommendationssectionattheendofthisdocument).


LitterBehaviorandDemographicVariables

Analysisofvariancewasusedtolookatlitteringbehaviorgroupdifferencesforgenderandeducation.Table18displaysthemeansandstandarddeviationsforthesevariables.

Table18.GroupMeansofLitteringBehavior(010)byGender,Education,TypeofResidence,andTypeofVehicle

Frequency Mean SDGender(IC1)** Female 677 0.49 1.28Male 358 0.83 1.70 1035 0.61 1.45EducationalAttainment(QD4) LessThanHighSchool 61 0.78 1.74HighSchoolGraduate 562 0.66 1.57CollegeGraduate 386 0.47 1.11 1009 0.60 1.42TypeofResidence(QD6) Housesingledetached 799 0.56 1.38AptorCondo 157 0.69 1.67Mobilehome 54 0.97 1.84 1010 0.61 1.45TypeofVehicle(QD7) Car 571 0.57 1.45Pick‐UpTruck 126 0.91 1.96VanorMini‐Van 104 0.64 1.42SUV 151 0.53 1.04Don'tDrive 51 0.41 1.13Other 5 0.46 0.79 1008 0.60 1.46Note:Frequency=numberofoccurrences,Mean=average,SD=standarddeviation.**differencesignificantatthep<.01level

Thesedatashowsignificantdifferencesbetweenmalesandfemalesforself‐reportedlitteringbehavior.Malesaremuchmorelikelytoreportlitteringbehaviorthanarewomen.Thisfindingisconsistentwithotherresearchonself‐reportedlitteringbehavior.Thesedatashownosignificantdifferencesforlitteringbehavioracrosseducationallevels,typeofresidence,ortypeofvehicle.Althoughthedataindicateariseinlitteringbehavioraseducationalleveldrops,thedifferencesarenotstatisticallysignificant.


PredictingLitteringBehavior

Thissurveywasdesignedtounderstandthevariablesassociatedwithlitteringbehavior.AMultipleRegressionAnalysiswasperformedtopredicttheimpactofselectedvariablesonlitteringbehavior.Thisanalysisshowstheeffectofthisgroupofpredictorvariablesonlitteringbehavior.Table19displaystheunstandardizedregressioncoefficients(b)forthepredictorvariables.

Table19.UnstandardizedRegressionCoefficientsfromaMultipleRegressionAnalysis:LitteringBehaviorby

PredictorVariables

b Age ‐0.010 **Gender 0.196 *LevelofEducation ‐0.021 CurrentSmoker 0.234 *ImportanceofLitter ‐0.047 **HeardorSeenLitterMessage 0.141 Livabilityquotient 0.015 PersonalNormsScale ‐0.019 MotivatorsandBarriersScale 0.107 **

N=881;*p<.05,**p<.01

Ninepredictorvariableswerechosentogaugetheimpactoftheirpresenceonself‐reportedlitteringbehavior.Fivevariablesemergedaspredictorsoflitteringbehavior.Age,gender,currentsmoker,importanceoflitter,andthemotivatorsandbarriersscaleallshowstatisticallysignificantrelationshipstoself‐reportedlitteringbehavior.

Thecurrentsmokervariablewasdummycodedinordertoincludeitinthemodel(smoker=1,non‐smoker=0).Thecorrelationcoefficientindicatesthatinthecontextofalltheothervariablesinthemodel,currentsmokerstatushasthegreatestimpactonself‐reportedlitteringbehavior.Therelationshipbetweenageandlitteringbehaviorisnegative;asageincreases,litteringbehaviordecreases.Thevariableforgenderwasdummycodedinordertoincludeitinthemodel(male=1,female=0);therefore,thepositivecorrelationcoefficientindicateshigherself‐reportedlitteringratesamongmen.Thereisalsoanegativerelationshipbetweentheimportanceoflitterinyourtownandlitteringbehavior;astheperceivedimportanceoflitterasanissueincreases,litteringbehaviordecreases.Finally,theanalysisshowsarelationshipbetweenthemotivatorsandbarriersscale(those


likelytobemotivatedtolitterbecauseofmoodorlackoftrashcans)andlitteringbehavior.

Conclusion

Inthecontextofthisgroupofpredictorvariables,currentsmokerstatusemergedasanimportantpredictorofself‐reportedlitteringbehavior.Although,thelitteringbehaviorvariableincludesgenerallitteraswellascigarettebuttlitter,thisfindingisconsistentwithlandsurveylitterstudiesthatshowcigarettebuttsasthemostlittereditemintheUnitedStates.Theeffectofageonlitteringbehaviorisconsistentwithotherresearchresultswhichshowthatasageincreases,litteringbehaviordecreases.Themodelshowedthatmenaremorelikelytoreportlitteringthanarewomen.Pastresearch,aswellastheresultsofourcurrentobservationalresearchsuggeststhattherearenoobservedlitteringbehaviordifferencesbetweenmenandwomen.Thedifferencesmaterializeintheself‐reportedbehaviors.Theperceivedimportanceoflitterasanissueshowsanimpactonlitteringbehavior;thoseindicatingagreaterperceivedlevelofimportanceoflitterasanissue,displaylowerlevelsoflitteringbehavior.Themotivatorsandbarriersscalealsohelpedtoexplainlitteringbehavior.Respondentswhoweremorelikelytoagreewiththebarriersstatementsweremorelikelytoreportlitteringbehavior.Inotherwords,themorelikelyoneistoreportthatbeinginabadmoodortheabsenceofatrashcaninducesthemtoimproperlydisposeoflitter,themorelikelytheyaretoreportlitteringbehavior.(Reliabilityanalysisandbivariatecorrelationswereconductedforbothscalestoinsurethevariableswerenotmeasuringthesameconstructs;whilereliabilitiesforbothscalesexceeded.9,thecorrelationbetweenthetwovariableswasonlymodest,r=.312).

TheImportanceofLitter

Analysisofvariancewasusedtoexploretherelationshipbetweentheperceivedimportanceoflitterinthetownorcity(QM1),andcommunityappearance(QBI).Table20displaysthegroupmeansfortheimportanceoflitteringinyourtownbyeachcommunitylivabilityitem.

Table20.GroupMeansforImportanceofLitteringinYourTown(010)byCommunityLivabilityItems

Frequency Mean SDVisibleLitter(QBI1)**

1Virtuallynovisiblelitter 342 8.42 2.722Uponcarefulinspectionasmallamountoflitterisobvious 507 8.04 2.433Visiblelittercanbeseenthroughoutthearea 125 7.48 2.714Majorillegaldumpsitesarepresent 21 7.33 3.40Total 995 8.09 2.61


Frequency Mean SDHowCleanisCommunity(QBI2)**

1Veryclean 535 8.49 2.462Somewhatclean 409 7.72 2.563Notveryclean 41 7.00 3.354Notatallclean 8 5.13 4.52Total 993 8.08 2.61StreetsandSidewalksMaintained(QBI3)**

1Verywellmaintained 499 8.52 2.332Somewhatmaintained 296 7.65 2.683Notverywellmaintained 81 7.57 2.844Notatallmaintained 43 7.19 3.24Total 919 8.09 2.58StreetsandSidewalksSafeDuringDay(QBI4)**

1Verysafe 647 8.23 2.492Somewhatsafe 227 7.89 2.793Somewhatunsafe 45 6.93 3.014Veryunsafe 20 8.20 2.67Total 939 8.08 2.61StreetsandSidewalksInvitingQBI5**

1Veryinviting 445 8.64 2.252Somewhatinviting 356 7.62 2.723Somewhatuninviting 57 6.84 3.124Veryuninviting 52 6.96 3.25Total 910 8.03 2.63NumberofFlowers,Plants,andTreesQBI6**

1Therearealotofflowers,plants,andtrees 535 8.38 2.392Therearesomeflowers,plants,andtrees 349 7.82 2.743Therearenotmanyflowers,plants,andtrees 79 7.49 2.764Therearenoflowers,plants,andtrees 29 7.90 3.30Total 992 8.10 2.59AttractivenessofFlowers,Plants,andTrees(QBI7)**

1Theyareveryattractive 532 8.45 2.442Theyaresomewhatattractive 370 7.88 2.523Theyaresomewhatunattractive 56 7.20 2.964Theyareveryunattractive 25 6.12 3.91Total 983 8.10 2.59AttractivenessofInfrastructure(QBI8)**

1Theyareveryattractive 310 8.63 2.312Theyaresomewhatattractive 471 7.93 2.553Theyaresomewhatunattractive 113 7.73 2.644Theyareveryunattractive 38 7.21 3.60Total 932 8.11 2.56Note:Frequency=numberofoccurrences.Mean=average.SD=standarddeviation.**indicatesstatisticalsignificanceatthep<.01level.


Allrelationshipsbetweentheimportanceoflitterinyourtownandthecommunitylivabilityitemswerestatisticallysignificant.Theanalysisshowedthatthegroupsthatreportedmorelitterintheircommunity(QBI1),alsoreportedthatlitterwaslessimportant.Thesamewastrueforthehowcleanisyourcommunityitem(QBI2);thosegroupsreportingtheylivedincommunitiesthatwerenotverycleanandnotatallclean,statedthatlitterwaslessimportant.Respondentswhoreportedtheirstreetsandsidewalksweresomewhatmaintained,notverywellmaintained,andnotatallmaintained(QBI3)indicatedthatlitterwaslessimportantintheirtownthanthosewhoreportedtheirstreetsandsidewalkswereverywellmaintained.

Thesafetyofstreetsandsidewalksduringdaylighthoursvariable(QBI4)showedamuchdifferentpattern;thosewhoindicatedtheirstreetswereverysafeandthosewhoindicatedtheirstreetswereveryunsafereportedlitterbeingofgreaterimportanceintheirtownsthanrespondentsintheothertwogroups(somewhatsafe,orsomewhatunsafe).Forrespondentsstatingtheirstreetsandsidewalkswereaveryinvitingplaceorsomewhatinvitingplacetowalk(QBI5),theimportanceoflitterintheirtownwasgreaterthanforthosewhostatedtheirstreetsandsidewalksweresomewhatinvitingorveryuninviting.Respondentswhoindicatedtheircommunityhadalotofflowers,plants,andtrees(QBI6),weremorelikelytostatethatlitterwasanimportantissueintheircommunitythantheotherthreegroups.Thosewhoindicatedtheflowers,plants,andtreesintheircommunitywereveryattractive(QBI7)weremorelikelytostatethatlitterwasanimportantissueintheirtown.Thesamepatternexistedfortheattractivenessoftheinfrastructure(QBI8);groupswhoindicatedthecommunityinfrastructurewasveryattractivereportedagreaterlevelofimportance,comparedtothosewhoindicatedthattheircommunityinfrastructurewasveryunattractive.

CONCLUSIONS

Theoverallpatternofrelationshipsfortheseitemsisquiteinteresting.Thepatternindicatestherearesignificantdifferencesbetweengroupsofpeoplewholiveintownsorcitieswithlittlevisiblelitter,withattractivelandscapingandinfrastructure,andwithstreetsandsidewalksthatarepleasanttowalk,andthosegroupswhodonot.Thegroupsthatliveinmoreattractivesurroundingsperceivetheissueoflitterasmoreimportantintheirtownorcitythanthosewholiveinlessattractivesurroundings.Thisisnottosaythattheindividualdoesnotperceivelittertobeanissue.Morelikely,thepresenceoflitterandlessattractivesurroundingsisanindicatorthatlitterisnotanissueintheirtownorcity.Inotherwords,iflitterwereanissue,thanitwouldnotbepresent,thetownwouldbemoreattractive,andperhapsamorepleasantplaceinwhichtowalk.


PARTV:CONCLUSIONSANDRECOMMENDATIONS

Inthisreport,wehavesummarizedthefindingsfromthreemethodologicalapproachestostudyinglitteringbehavior.Inthefirst,weconductedsystemicobservationsof9,757individualsrandomlyselectedfrom130sitesacrossthecountry.Inthesecond,wereportfindingsfromasmallsampleofinterceptinterviewsconductedwith102individualsat15sites.Inthethirdstudy,wereportfindingsfromanationaltelephonesurveyof1,039households.Eachofthesethreeapproachesprovidesauniqueperspectiveonlitteringbehavior,andinthisfinalsection,wewanttostepbackanddrawsomebroadconclusionsandrecommendations.

LitteringRates

Oneofthefirstquestionsthatemergesindiscussionsoflitteringis:“howoftendopeopledoit?”Inthe1968survey,resultsshowed“exactlyhalf”oftherespondentsinanationalsurveyreportedatleastonelitteringbehavior(among8scenarios).Whilethedetailsfromthis40‐year‐oldstudyarenolongeravailable,the50%figureisameaningfulmark.

Ourcurrentdataindicatethatlitteringrateshavedeclineddramaticallyinthepast40years.Inourtelephonesurvey,wherewereplicatedascloselyaspossiblethemeasurementprocedurefrom1968,wefoundthatonly15%ofrespondentsreportalitteringactinthepastmonth(usingthesame8KABscenarios).Andacrossarangeofdifferentlocationsandlittertypes,wefindthatAmericansreportaverylowfrequencyoflittering.

However,it’simportanttopointoutthelimitationsofself‐reportsurveys.Giventhatpeoplealmostunanimouslyviewlitteringasabadthing,it’snotsurprisingthatveryfewreportdoingit.Inadditiontotheproblemofself‐reportbias,itmayalsobethecasethatlitteringhappensbyaccident.Inthiscase,respondentswouldbereportingcorrectly,butstillunderestimatingtheamountthattheylitter.

Toaddresstheselimitations,weconductedaseriesofobservationswitharandomsampleofindividualsacrossthecountry.Ourresultsindicatethat17%ofalldisposalsinthepublicspacesweobservedwerelitter.Inaddition,4%ofallindividualsthatpassedthroughthepublicspacesweobservedlittered.It’salsoclearfromourobservationsthatmostoftheseactsoflitteringarenotbyaccident,andalargemajority(81%)ofallobservedactsoflitteringoccurredwithclearlynotableintent.

Intermsoflittereditems,ourresultsshowthatthemostfrequentlylittereditemsarealsothosemostcommonlyfoundontheground.Whileourmethodology


doesnotallowsolidconclusionsaboutthepercentageoflitteronthegroundthatoriginatesfromindividualsdroppingorthrowinganitem(asopposedtotrashcansblowingover,litterblowingfromuncoveredloads,oraninefficienttrashcollectionsystem),itdoesseemclearthatalargemajorityoflitteroriginatesfromtheindividualsthatpassthroughapublicspace.Themostfrequentlyobserveditemsthatwerelitteredincluded:cigarettebutts,foodremnants,andfoodwrappers.

Themethodsthatwehavedevelopedandreportedwereintendedtobe“replicable.”Thatis,wehaveattemptedtobuildanobservationalmethodologythatcanbeimplementedovertimeandinarangeoflocations.Inadoptingthesamemethodology,itwillbepossibletodrawsolidconclusionsaboutchangeovertime,toinformbothlocalandnationallitter‐preventionprograms,andtoevaluatethesuccessoflitterpreventionstrategies.

UnderstandingLitteringBehavior

Theresultsreportedinthisdocumentshowthatasof2008,litteringbehaviorremainsaseriousnationalproblem.Americansviewlitterasaseriousissue;manyindividualsfeelapersonalobligationnottolitter;andtheywanttoliveinclean,litter‐freecommunities.Butyetthebehaviorpersists.

Theuniqueresearchdesigndevelopedforournationalobservationsdrawsonthequantitativetechniqueofmulti‐levelmodeling.Byobservingrandomsamplesofindividualswithinalocation,andmultiplelocationsacrossthecountry,weareabletoascertaintheamountofvariabilityinlitteringbehaviorthatisduetotheperson,orduetothecontext.Forgenerallittering,wefindthat15%ofthevarianceinlitteringbehaviorresultsfromcontextualconstraints(e.g.,insufficientnumberofreceptacles,existinglitterontheground),whiletheremainderresultsfromtheindividual(e.g.,lackofmotivation).Forcigarettebuttlitter,wefindthatmorethantwiceasmuchvarianceinlitteringbehavior(38%)isduetothecontextualdemands,and62%isduetotheindividual.

Withregardtothesituationalfactorsthatcontributetolittering,twoconsistentfindingsemergedfromourobservations.Firstistheavailabilityoftrashreceptacles.Whileourdatashowthatalmostallpublicplaceshaveatleastonetrashreceptacle(upfrom56%reportedinthe1968KABstudy),thenumberandconvenienceofreceptaclesstronglyinfluenceslitteringbehavior.Whilewecertainlyobservedtherareexceptionofanindividuallitteringrightnexttoareceptacle,theaveragelitteringbehavioroccurredatadistanceof29feetfromthenearestreceptacle.Inaddition,theamountoflitteralreadypresentinasitestronglyinfluencedtheprobabilitythatapersonwouldlitter.

Withregardtoindividualfactorsthatcontributetolittering,wewanttohighlighttwogeneralfindings.Thefirstisage.Consistentlyacrossallthreeapproaches(observation,intercept,andsurvey),wefindthatolderindividualsare


lesslikelytolitterthanyounger.Thisfindingechoestheconclusionmadeinthe1968report.Thesecondpredictivefactorisaperson’sbeliefintheimportanceoflitter.Inthetelephonesurvey,thoseindividualswhobelievedthatlitterwasamoreimportantissuewerelesslikelytoreportlittering.Inaddition,ourobservationandinterceptdatashowthatindividualswhofeelapersonalobligationnottolitterwerelesslikelytodoso.

Anotherfindingofinterestpertainstogender.Whilemuchhasbeensaidaboutthehighrateoflitteringbehavioramongmales(particularlyyoungmales),ourdatapresentaconflictingstory.Whileitisclearlythecasethatyoungmalesaremorelikelytoreportlittering,andtoreportalowerpersonalobligationtonotlitter,ourobservationsfailedtoshowanyoverallgendereffectsinobservedlitteringrates.Tous,thissuggeststhatthegendereffectismoreaproductofself‐reportbiasthanbehavioraldifferences,andthatoutreacheffortsshouldtargetbothmenandwomen.

LitterPrevention

Thefindingsfromourresearchpointtoseveralstrategiesforpreventinglitter.Thesestrategiesincludeacombinationofbothstructuralandmotivationalactivities.Wepresentthreestrategiesbelowthatareconsistentwithourresearchfindings.Butwedonotproposethisasanexhaustivelist,andweencouragethereadertothinkcreativelyaboutwaystolinkthereportedfindingstolitterpreventionstrategies.

1.Beautification(andcleanup).Acrossourdataweconsistentlyfoundthatlitteredenvironmentsattractmorelitter.Thisfindingisnotnew,andindeeditwasnotedintheresearchfindings40yearsago.Inessence,individualsuseavarietyofcuesfromtheirsurroundingenvironmenttodeterminewhatiscommonandacceptedbehavior.Thepresenceoflittercommunicatesthenormforthatsituation,andtheacceptabilityoflittering.Inaddition,theexistinglitterwillrequireclean‐up,soonemorepiecewon’tmatter.

Tothisend,werecommendconsistentandongoingcommunityclean‐upactivities.Reducingtheamountofexistinglitterinalocationisasurefirewaytoreducetherateoflitteringbehavior.

However,wealsorecommendgoingbeyondcommunityclean‐ups.Whilesucheffortsarelaudableintheirgoalsandimpact,theoutcomesarelikelytobeshortlived.Clean‐upeffortsdealwiththesymptoms,butnottheunderlyingproblem.Inadditiontothe“socialnorms”aspectofexistinglitter,wealsofoundsignificanteffectsforphysicalappearance.Throughoutourstudies,wefoundconsistentevidencethatcommunitiesthatmakeaneffortto“beautify,”eitherthroughlandscaping,hardscape,maintenance,andcleanliness,generallyresultinlowerratesoflitteringbehavior.Inpart,anattractivecommunitycommunicatesasocial


norm,butitalsopromotesasenseofresponsibilityandapersonalobligationnottolitter.Communitiesandcitiesthatexpectindividualstofeelapersonalresponsibilitytonotlitter(andtocareabouttheappearanceofthelocalcommunity)shouldsettheexample.Beautificationeffortssetastandardforthecommunity,andmodeltheresponsiblebehaviorsthatcommunityleadershopetoengenderamongresidents.Indeed,clean‐upeffortsintheabsenceofinfrastructurechangesarelikelytoproduceeffectsthatareshort‐lived.

2.BehavioralAffordances.Relatedtoourrecommendationtosupportbeautificationefforts(above),wealsofindconsistentevidencefortheimportanceofopportunity.Thatis,thecontextshouldprovideaconvenientandaccessiblemeansforproperdisposaloftrashandlitter.Whileourdatashowalargeincreaseintheavailabilityofreceptaclesinpublicplacesoverthepast40years,wealsofoundthatdistancetoatrashcanwasastrongpredictoroflitteringbehavior.Providingeasily‐identifiableandaccessiblereceptaclescangoalongwaytowardreducinglitteringrates.Thisrecommendationechoesthatmadeinthe1968KABreport,whichstates:

…theobviousfactmustbekeptinmindthatnomatterhow“sold”apersonmaybeonnotlittering,ifhecanfindnoplacetodisposeofhislitterhewillalmostcertainlydiscarditwhereveritisconvenientandnottooconspicuous.

Theissueofbehavioralaffordancesisespeciallyimportantforcigarettebutts.Ourobservationaldatasuggestthatdisposalofcigarettebuttsismorestronglyclusteredwithinlocations,andyetlessthanhalf(47%)ofthelocationswevisitedprovidedanashreceptacle.Giventheincreaseinlegislationregardingindoorsmoking,anincreasingnumberofsmokersaremovingoutsidetosmoke.However,theinfrastructureforcollectingashesandlitcigarettesiswoefullybehindthesepolicies,andourdatasuggestthatmoreeffortstoaffordsmokersanopportunityforproperdisposalareneeded.

3.AwarenessandMotivationCampaigns.Inadditiontoourrecommendationsforbeautificationandinfrastructure,wecontinuetoseeanimportantroleforawarenessandmotivationcampaigns.Inourobservationaldata,wefindthat85%ofthevarianceingenerallittering,and62%ofthevarianceincigarettebuttlitteringresultsfromindividualdifferences.Theseincludedemographic(forexample,income,age,gender),attitudinal(forexample,lackofconcernaboutlitter),andmotivationaldifferences,andtheyspeaktotheimportanceofaconsistentandongoingmessageabouttheimportanceandreasonsfornotlittering.

However,wewanttobeginbysuggestingwhatnottodo.Toooften,awarenesscampaignsconveymessagesaboutahighlitteringrate,ordepictimagesofhighlylitteredenvironments.Whilewellintentioned,suchmessagesconveyanormativemessagethatotherpeoplelitter.Andconsistentlyacrossourresearch(andinotherpublishedstudies),normativemessagesaboutthehighrateoflittering


canincreaselitteringrates.Asweobservedrepeatedlyacrossthecountry:litterbegetslittering.Messagesthatshowlitteringascommon(i.e.,normative)makeitacceptable,andwillgenerallyunderminetheultimategoalofreducinglitter.Westronglyadviseagainstsuchmessages.Instead,weadvocateformessagesthathighlightthedramaticdeclineinlitteringratesoverthepast40years,thegenerallyinfrequentoveralllitteringrate,andthewidespreadbeliefthatlitteringiswrong.

Acrossourresearch,wefindthenearunanimousbeliefthatlitteringiswrong.This,coupledwithotherresearchontheroleofinjunctiveandpersonalnorms,suggestsapromisingavenueformotivationalmessages.Withregardtopersonalnorms,wefindstrongevidenceinboththebehavioralobservationsandtelephonesurveysthatindividualswhofeelastrongpersonalobligationnottolitterarelesslikelytodoso,eveninsituationsthatdonotaffordopportunitiesforproperdisposal.Inaddition,priorworkhasshownthataninjunctivenormofsocialdisapprovalcanprovideastrongbehavioralmotivation.Inshort,onlyafewdeviantindividualswillcontinuetolitter,andtheseindividualsaredisapprovedofbythemajority(seeCialdini,2003;Schultzetal.,2008).

Finally,wewanttoreiteratetheimportanceoffocusingonthelocalcommunity,andlocalcontexts.Allofourobservationsfocusedonpublic,outdoorspaces,andtherewasaconsiderabledegreeofvariabilityacrossthecountry.Whileourfindingsandrecommendationswillbeusefultostructureanational‐levelprogramandoutreachcampaign,wewanttoemphasizethatlitteringisalocalissue.Whilethenational‐levelagendacanhelptoguideandstructurelitter‐preventionefforts,theseactivitiesneedtobeimplementedlocally.Assuch,weurgelocalorganizationstounderstandthemotivationalandstructuralbarriersthatexistwithintheircommunities,andtodeviseoutreachandinterventionstrategiesthataretailoredtomeettheirneeds.Thedatainthisreportcanprovideastartingpoint,andthenational‐levelfindingscanhelptoframeandbringfocustothework.Butitislocalorganizations,communities,andultimatelyindividuals,thatwillbringanendtolitter.


REFERENCES

CentersforDiseaseControlandPrevention(2007).Adultswhoarecurrentsmokers.Behavioralriskfactorsurveillancesystemmap.RetrievedAugust23,2007fromhttp://apps.nccd.cdc.gov/gisbrfss/select_question.aspx.

Cialdini,R.B.(2003).Craftingnormativemessagestoprojecttheenvironment.CurrentDirectionsinPsychologicalScience,12,105‐109.

Cialdini,R.B.,Barrett,D.W.,Bator,R.,Demaine,L.,Sagarian,B.J.,Rhoads,K.,&Winter,P.L.(2005).Activatingandaligningsocialnormsforpersuasiveimpact.Manuscriptsubmittedforpublication.

Cialdini,R.B.,Kallgren,C.A.,&Reno,R.R.(1991).Afocustheoryofnormativeconduct:Atheoreticalrefinementandreevaluationoftheroleofnormsinhumanbehavior.AdvancesinExperimentalSocialPsychology,21,201‐234.

Cialdini,R.B.,Reno,R.R.,&Kallgren,C.A.(1990).Afocustheoryofnormativeconduct:Recyclinginconceptofnormstoreducelitteringinpublicplaces.JournalofPersonalityandSocialPsychology,58(6),1015‐1026.

Cope,J.G.,Huffman,K.T.,Allred,L.J.,&Grossnickle,W.F.(1993).Behavioralstrategiestoreducecigarettelitter.JournalofSocialBehaviorandPersonality,8,607‐619.

DepartmentofEnvironmentandConservation(NSW).(2004).NewSouthWaleslitterreport2004.RetrievedMarch15,2007fromhttp://www.epa.nsw.gov.au/litter/research.htm.

Davis,J.A.,Smith,T.W.,&Marsden,P.V.GENERALSOCIALSURVEYS,1972‐2006[CUMULATIVEFILE][Computerfile].ICPSR04697‐v2.Chicago,IL:NationalOpinionResearchCenter[producer],2007.Storrs,CT:RoperCenterforPublicOpinionResearch,UniversityofConnecticut/AnnArbor,MI:Inter‐universityConsortiumforPoliticalandSocialResearch[distributors],2007‐09‐10.

Durdan,C.A.,Reeder,G.D.,&Hecht,P.R.(1985).Litterinauniversitycafeteria:Demographicdataandtheuseofpromptsasaninterventionstrategy.EnvironmentandBehavior,16,387‐404


Finnie,W.C.(1973).Fieldexperimentsinlittercontrol.EnvironmentandBehavior,5(2),123‐144.

Geller,E.S.(1973).Promptinganti‐litterbehaviors.Proceedingsofthe81stAnnualConventionoftheAmericanPsychologicalAssociation,8,901‐902.

Geller,E.S.(1975).Increasingdesiredwastedisposalswithinstructions.ManEnvironmentSystems,5,125‐128.

Geller,E.S.,Brasted,W.,&Mann,M.(1980).Wastereceptacledesignsandinterventionsforlittercontrol.JournalofEnvironmentalSystems,9,145‐160.

Geller,E.S.,Winett,R.A.,&Everett,PB.(1982).Preservingtheenvironment:Newstrategiesforbehaviorchange.Elmsford,NY:Pergamon.

Geller,E.S.,Witmer,J.F.,&Orebaugh,A.L.(1976).Instructionsasadeterminantofpaper‐disposalbehaviors.EnvironmentandBehavior,8(3),417‐493.

Geller,E.S.,Witmer,J.F.,&Tuso,M.A.(1977).Environmentalinterventionsforlittercontrol.JournalofAppliedPsychology,62(3),344‐351.

Grasmick,H.,BursikR.,&Kinsey,K.(1991).Shameandembarrassmentasdeterrentstononcompliancewiththelaw:Thecaseofanantilitteringcampaign.EnvironmentandBehavior,23(2),233‐251.

Herberleiin,T.A.(1971).Moralnorms,threatenedsanctionsandlitteringbehavior.Unpublisheddoctoraldissertations,UniversityofWisconsin,Madison.

Kallgren,C.A.,Reno,R.R.,&Cialdini,R.B.(2000).Afocustheoryofnormativeconduct:Whennormsdoanddonotaffectbehavior.PersonalityandSocialPsychologyBulletin,26,1002‐1012.

KeepAmericaBeautiful(2007).KAB’ssevenprimarysourcesoflitter.RetrievedSeptember1,2007fromhttp://www.kab.org/site/PageServer?pagename=litter_sevensources.

Krauss,R.M.,Freedman,J.L.,&Whitcup,M.(1978).Fieldandlaboratorystudiesoflittering.JournalofExperimentalSocialPsychology,14,109‐122.

Meeker,F.L.(1997).Acomparisonoftable‐litteringbehaviorintwosettings:Acaseforacontextualresearchstrategy.JournalofEnvironmentalPsychology,17,59‐68.


Nolan,J.,Schultz,P.W.,&Knowles,E.(inpress).Usingpublicserviceannouncementstochangebehavior:Nomoremoneyandoildownthedrain.JournalofAppliedSocialPsychology.

Reiter,S.M.,&Samuel,W.(1980).Litteringasafunctionofpriorlitterandthepresenceorabsenceofprohibitivesigns.JournalofAppliedSocialPsychology,10,45‐55.

R.W.Beck(2007).LiteratureReview–Litter:Areviewoflitterstudies,attitudessurveysandotherlitterrelatedliterature.RetrievedSeptember1,2007fromhttp://www.kab.org/site/DocServer/Litter_Literature_Review.pdf?docID=481.

Sansone,C.,&Harackiewicz,J.(2000).Intrinsicandextrinsicmotivation:Thesearchforoptimalmotivationandperformance.SanDiego,CA:AcademicPress.

Schultz,P.W.,Tabanico,J.,&Rendón,T.(2008).Normativebeliefsasagentsofinfluence:Basicprocessesandreal‐worldapplications.InR.Prislin&W.Crano(Eds.),Attitudesandattitudechange(pp.385‐409).NewYork:PsychologyPress.

Schultz,P.W.,&Tabanico,J.(2008).Community‐basedsocialmarketingandbehaviorchange.InA.Cabaniss(Ed.),Handbookonhouseholdhazardouswaste(pp.133‐157).Lanham,MD:GovernmentInstitutesPress.

Sibley,C.G.,&Liu,J.H.(2003).Differentiatingactiveandpassivelittering:Atwo‐stageprocessmodeloflitteringbehaviorinpublic.EnvironmentandBehavior,35,415‐433.

Williams,E.,Curnow,R.&Streker,P.(1997).UnderstandinglitteringbehaviourinAustralia.Abeverageindustryenvironmentcouncilpublication:Acommunitychangeconsultantsreport.Victoria.


APPENDIXA:REFERENCETABLES

Table21.SitesbyLocationandObservationType

State City CityType SiteType ObservationType

NumberofObservations

AR LittleRock Urban CityCenter General 160 Bar/Restaurant Smoking 7 Recreation General 95 RestStop General 155 GasStation General 165 NorthLittleRock Suburban CityCenter General 192 Recreation General 37 Medical/Hospital Smoking 7 GasStation General 94 Maumelle Rural Recreation General 68CA SanDiego Urban CityCenter General 150 FastFood General 78 Recreation General 78 Recreation Smoking 15 Carlsbad Suburban CityCenter General 193 FastFood General 89 Bar/Restaurant Smoking 15 Recreation General 177 Recreation Smoking 19 RestStop General 207 Ramona Rural FastFood General 130 RestStop General 207 Retail Smoking 22 GasStation General 172GA Atlanta Urban CityCenter General 140 FastFood General 138 RestStop General 139 Retail Smoking 29 Marietta Suburban CityCenter General 224 Bar/Restaurant Smoking 6 GasStation General 189




Cartersville Rural

FastFood General 69 Recreation General 147 Recreation Smoking 7 GasStation General 73IL Chicago Urban FastFood General 187 Medical/Hospital Smoking 42 Retail Smoking 118 Elgin Suburban FastFood General 110 RestStop General 138 FoxRiverGrove Rural CityCenter General 56 FastFood General 269 Bar/Restaurant Smoking 24 Recreation General 43 Recreation Smoking 9KY Louisville Urban CityCenter General 78 RestStop General 173 Medical/Hospital Smoking 58 Jeffersontown Suburban FastFood General 131 Recreation Smoking 9 Retail Smoking 28 GasStation General 167 Prospect Rural Retail Smoking 12NV LasVegas Urban CityCenter General 235 FastFood General 79 Recreation Smoking 10 Retail Smoking 52 Henderson Suburban CityCenter General 272 Recreation General 91 BoulderCity Rural FastFood General 159 RestStop General 108 Retail Smoking 36 GasStation General 196




NM

Albuquerque Urban CityCenter General 316 GasStation General 101 ParadiseHills Suburban CityCenter Smoking 30 Medical/Hospital Smoking 31 LosRanchos Rural Recreation Smoking 4 RestStop General 73NY Albany Urban CityCenter General 96 Recreation General 72 Medical/Hospital Smoking 34 Plattsburgh Suburban FastFood General 70 Bar/Restaurant Smoking 35 Retail Smoking 44 LakePlacid Rural Recreation General 85UT SaltLakeCity Urban FastFood General 117 Bar/Restaurant Smoking 4 Recreation General 78 Recreation Smoking 5 RestStop General 93 Medical/Hospital Smoking 32 WestValley Suburban CityCenter General 182 FastFood General 203 Recreation General 74 BountifulCity Rural CityCenter General 156 FastFood General 140 Recreation General 79 Retail Smoking 13 GasStation General 287VT Burlington Urban CityCenter General 119 FastFood General 63 Recreation General 67 Recreation Smoking 4




SouthBurlington

Suburban

FastFood General 108 Bar/Restaurant Smoking 6 Williston Rural CityCenter General 79 FastFood General 154 Recreation General 15 RestStop General 105TOTAL 10 30 100 9,757


APPENDIXB:BEHAVIORALOBSERVATIONDOCUMENTS


GeneralLitterObservationBooklet


General Litter Observations

Book _____ of _____

Researcher:___________ Date:________________

Location:__________________________________

POST OBSERVATION SUMMARY (per booklet)

Total Observed

Total “Left with No Visible Object”

Total “Left Site with Object

Total Disposers

Comments/Concerns/ProblemsEncountered


Object Disposed Disposal Method Litter Strategy

1 Beverage Bottle: Plastic

2 Beverage Bottle: Glass

3 Beverage Can

4 Beverage Cup

5 Food Wrapper

6 Food Container

7 Food Remnants

8 Cigarette Butt (21+ only)

9 Paper

10 Napkin/Tissue

11 Plastic Bag

12 Combo/Mixed Trash

13 Other:_____________

1 Pocketed

2 Trash Can

3 Recycling Bin: Correct

4 Recycling Bin: Incorrect

5 Ashtray

6 Separated

7 Ground

8 Planter

9 Bushes/Shrubbery

10 On/Around Receptacle

11 Other:___________

0 N/A

1 Wedge

2 Flick/Fling

3 Shoot & Miss

4 Bury

5 Drop: Intent

6 Drop: No Intent

7 Inch Away

8 Sweep

9 90%

10 Other:_______________

Receptacles: At Start Receptacles: At Disposal

Nearest Trash Can: _______feet

Nearest Recycling Bin: _______feet

Nearest Ashtray: _______feet

Nearest Trash Can: _______feet

Nearest Recycling Bin: _______feet

Nearest Ashtray: _______feet

Activity Before Disposal Activity After Disposal Describe: Describe:

Group Setting at Disposal Randomization Sheet 0 Alone

1 In Group # = ________

2 Nearby # = ________

1 2 3 4 5 6

Notes Survey Sites Only Anything unusual? Other Enforcements? Selected for Intercept?

1 YES 2 NO Survey Result: MISSED REFUSED COMPLETED


CigaretteButtLitterObservationBooklet


Cigarette Litter Observations

Book _____ of _____

Researcher:___________ Date:________________

Location:__________________________________

POST OBSERVATION SUMMARY (per booklet)

Total Smokers Observed

Total “Left Site” or “Unknown”

Total Disposers Observed

(Observed-Left Site/Unknown)

Comments/Concerns/ProblemsEncountered


Time Sunset

______ : ______ AM PM Before Sunset After Sunset

Gender Age (21+)

1 Male 2 Female 3 Unknown # = ___ ___

Receptacles: At Start Receptacles: At Disposal

Nearest Trash Can: _____feet

Nearest Ashtray: _____feet

Nearest Trash Can: _____feet

Nearest Ashtray: _____feet Disposal Method Litter Strategy

0 Left Site with Cigarette

1 Unknown/Unseen

2 Pocketed

3 Trash Can

4 Ashtray

5 Recycling Bin

6 Field Strip

7 Ground

8 Planter

9 Bushes/Shrubbery

10 On/Around Receptacle

11 Other:________________

0 N/A

1 Wedge

2 Flick/Fling

3 Shoot & Miss

4 Bury

5 Drop

6 Stomp

7 Other:_______________

Activity Before Disposal Activity After DisposalDescribe: Describe:

Group Setting at Disposal Randomization Sheet 0 Alone

1 In Group # = ________

2 Nearby # = ________

1 2 3 4 5 6

Notes Survey Sites Only Anything unusual? Other Enforcements? Selected for Intercept?

1 YES 2 NO Survey Result: MISSED REFUSED COMPLETED


APPENDIXC:INTERCEPTSURVEYINSTRUMENT


TO BE COMPLETED BY INTERVIEWER: Start Time _____:_____am/pm Observation ID# _____Location: ID#: _______________________Interviewer ID#: __________

1. What brings you here today? ________________________________________ 2. In a typical month, how often do you come here? ________ 3. Using this scale from 0 to 10, where 0 equals “not at all important” and 10 equals “very important,” in this location how important is the issue of litter to you? _____ 4. Of the people who pass through this location, what percentage of the people do you think litter?

__________% 5. Have you seen or heard a specific message about litter prevention in the past year? Yes

No 6. Have you seen or heard a specific message about a scheduled community clean up in the past year?

Yes No 7. Have you seen or heard any messages discouraging cigarette butt litter in the past year? Yes

No 8. Have you seen or heard any messages simply telling people ‘do not litter’ in the past year? Yes

No 9. [IF 5-8 = NO, SKIP TO Q10] Think of the most recent litter prevention message you’ve seen or heard in the past year. Where did you see it or hear it? (check one)

1. PSA on the radio 2. PSA/Commercial/News Report on Television 3. Billboard 4. Newspaper 5. Community Flyer/Bulletin 6. From a Friend/Family Member 7. Other________________________________________________

Next I am going to ask you a set of situational questions. There are no right or wrong answers. Please answer using this scale from 0 to 10 where 0 equals “no personal obligation” and 10 equals “very strong obligation.” Do you feel a personal obligation… _____10a. to not litter when you are holding an empty soft drink can and there are no trash cans

available? _____10b. to not litter when you are holding a gum wrapper and there are no trash cans available? _____10c. to stop and pick-up a piece of scrap paper that blows off a big stack of papers that you are

carrying in both arms? _____10d. to not litter when you are ill (fever, headache, muscle ache) and you would have to walk

out of your way to reach a trash receptacle?

_____10e. to pick-up a piece of paper you dropped when it is raining and you are getting soaked? 11. In the past month, have you littered? Yes No [SKIP TO Q12]


11a. Where did you litter? 11b. What did you litter? (Probe for Specific Litter Type) 11c. What would you say is the reason that you littered?

Now I would like to ask you a few questions about tobacco use…[X-OUT IF < 21 YRS)

12. Do you smoke cigarettes? Yes No [SKIP TO Q12] 12a.When you smoke outside, how do you usually dispose of the cigarette butt?

13. Do you consider cigarette butts to be litter? Yes No

For classification purposes, I’d like to ask you a few questions about yourself…

14. In what year were you born? __________ (21years old if born on or before 1987)

15. What is the highest grade of school you have completed?_____________________

That’s all the questions I have for you today. Thank you for sharing your thoughts.

Do you have any questions?

TO BE COMPLETED BY INTERVIEWER:

Interview End Time _____:_____am/pm

Gender: 1.Male 2.Female 3. Unknown

How well did the respondent understand the questions? 1.Very Well 2.Somewhat Well 3.Not Well

How attentive was the respondent? 1.Very Attentive 2.Somewhat Attentive 3.Not at all Attentive

Notes:


APPENDIXD:TELEPHONESURVEYINSTRUMENT


INTRO1. Hello, my name is _______ and I’m calling from Action Research, Inc., a public opinion research company. We are not selling anything. Your number was chosen randomly and I would like to ask you some questions about the health and well-being of your community. This survey will take less than 10 minutes to complete. You are one of 1,200 people across the nation who will be taking part in this important research project. INTRO2. Are you at least 18 years of age?

1. YES 2. NO [►ASK FOR HOUSEHOLD MEMBER AT LEAST 18 YEARS OF AGE]

INTRO3. We will not ask for your name or any other personal information that can identify you. The answers you give will be kept strictly confidential. You do not have to answer any questions you do not want to and you may stop the interview at any time. LIVABILITYQUOTIENT

TBI. Okay, let’s begin. For each of the following questions please choose the answer that best describes the physical environment in your community.

[RANDOMIZEORDEROFQUESTIONSINTHISSECTION]

QBI1.Howmuchlitterisvisibleinyourcommunity?

1. Virtually no visible litter 2. Upon careful inspection a small amount of litter is obvious 3. Visible litter can be seen throughout the area, likely requiring in an organized clean-up 4. Major illegal dumpsites are present, likely requiring equipment or extra manpower for

removal 8. DON’T KNOW 9. REFUSED

QBI2.Howcleanisyourcommunity?

1. Very clean 2. Somewhat clean 3. Not very clean 4. Not at all clean 8. DON’T KNOW 9. REFUSED


QBI3.Arethestreetsandsidewalksinyourcommunitywell‐maintainedforpeoplewhoarewalking?[CLARIFY,IFNECESSARY:“Iamaskingaboutyourcommunity,notnecessarilyyourspecificneighborhoodorareainfrontofyourhome.”IFRESPONDENTCONTINUESTOSTATEHE/SHEDOESN’THAVESIDEWALKS,CODEAS“NA.”]

1. Very well maintained 2. Somewhat maintained 3. Not very well maintained 4. Not at all maintained 7. NA – NO SIDEWALKS 8. DON’T KNOW 9. REFUSED

QBI4.Arethestreetsandsidewalksinyourcommunitysafeforpeopletowalkduringthedaylighthours?[CLARIFY,IFNECESSARY:“Iamaskingaboutyourcommunity,notnecessarilyyourspecificneighborhoodorareainfrontofyourhome.”IFRESPONDENTCONTINUESTOSTATEHE/SHEDOESN’THAVESIDEWALKS,CODEAS“NA.”]

1. Very safe 2. Somewhat safe 3. Somewhat unsafe 4. Very unsafe 7. NA – NO SIDEWALKS 8. DON’T KNOW 9. REFUSED

QBI5.Arethestreetsandsidewalksinyourcommunityaninvitingplaceforpeopletowalk?[CLARIFY,IFNECESSARY:“Iamaskingaboutyourcommunity,notnecessarilyyourspecificneighborhoodorareainfrontofyourhome.”IFRESPONDENTCONTINUESTOSTATEHE/SHEDOESN’THAVESIDEWALKS,CODEAS“NA.”]

1. Very inviting 2. Somewhat inviting 3. Somewhat uninviting 4. Very uninviting 7. NA – NO SIDEWALKS 8. DON’T KNOW 9. REFUSED


QBI6.Howwouldyoudescribethelandscapinginyourcommunity,forinstance,thenumberofflowers,plants,andtrees?

1. There are a lot of flowers, plants, and trees 2. There are some flowers, plants, and trees 3. There aren’t many flowers, plants, and trees 4. There are no flowers, plants, and trees 8. DON’T KNOW 9. REFUSED

QBI7.Howwouldyoudescribetheattractivenessoftheflowers,plants,andtreesinyourcommunity?

1. They are very attractive 2. They are somewhat attractive 3. They are somewhat unattractive 4. They are very unattractive 8. DON’T KNOW 9. REFUSED

QBI8.Howwouldyoudescribetheattractivenessofthingslike,benches,planters,streetsigns,andstreetlightsinyourcommunity?

1. They are very attractive 2. They are somewhat attractive 3. They are somewhat unattractive 4. They are very unattractive 8. DON’T KNOW 9. REFUSED

WORK ENVIRONMENT

TWE. Okay, I’d like to switch subjects for just a moment. QWE1. Do you currently work for pay?

1. YES 2. NO [►SKIP TO TTU] 8. DON’T KNOW [►SKIP TO TTU] 9. REFUSED [►SKIP TO TTU]

QWE2. Do you currently work primarily in an indoor or an outdoor environment?

1. INDOOR 2. OUTDOOR 8. DON’T KNOW 9. REFUSED


QWE3. Is that location designated as a smoke-free environment?

1. YES 2. NO 8. DON’T KNOW 9. REFUSED

CURRENT TOBACCO USE

TTU. Now I would like to ask you a few questions about tobacco use. QTU1. Have you smoked at least 100 cigarettes in your entire life?

1. YES 2. NO [►SKIP TO TKAB] 8. DON’T KNOW [►SKIP TO TKAB] 9. REFUSED [►SKIP TO TKAB]

QTU2. Do you smoke cigarettes every day, some days, or not at all?

1. EVERY DAY [►RECEIVES SMOKER QUESTIONS THROUGHOUT SURVEY] 2. SOME DAYS [►RECEIVES SMOKER QUESTIONS THROUGHOUT SURVEY] 3. NOT AT ALL [►SKIP TO TKAB] 8. DON’T KNOW [►SKIP TO TKAB] 9. REFUSED [►SKIP TO TKAB]

CIGARETTE DISPOSAL

QCD1. Does the location where you currently work have receptacles for cigarette butts?


QCD2. Does the vehicle you ride in most often have a receptacle for cigarette butts?


QCD3. Do you own a pocket ashtray?



LITTERINGBEHAVIOR–TYPE–KAB1968

TKAB.Forthenextsetofeightquestionspleasethinkaboutyourdailyactivitiesathome,atwork,whileshopping,andwhilesocializingwithfriendsandfamily.

Inthepastmonthhaveyou...

[RANDOMIZEQUESTIONS]

QKAB1. droppedgumorcandywrappersontheground,sidewalk,orstreet?

QKAB2. droppedfacialtissueontheground,sidewalk,orstreet?

QKAB3. droppedfoodorfoodwrappersonthestreetorhighway?

QKAB4. droppedbeerorsodacans,orbottlesonbeachesorotheroutdoorareas?

QKAB5. droppedpapercontainersofanykindontheground,sidewalk,orstreet?

QKAB6. leftanewspaperbehindforsomeoneelsetopickup?

QKAB7. leftpaper,foodremnants,orotherdiscardsatapicnicarea?

QKAB8. thrownthingsoutofacarorboatonthehighwaysorwaterways?

1. YES - - - 2. NO - - - 8. DON’T KNOW - - - 9. REFUSED

LITTERINGBEHAVIOR–TYPE/LOCATIONCOMPARISONS

TCOMP.Thenextsetofquestionsisaboutthetypesofitemscommonlyseenonoursidewalks,streets,inourparks,andonourhighways.

Usingascalefromzerototen,wherezeroequalsnotatalllikelyandtenequalsverylikelyhowlikelyareyoutodrop...

[RANDOMIZEBLOCKSOFQUESTIONS–BLOCK1,BLOCK2,BLOCK3,BLOCK4]

QCOMP1a. anapplecore,orabananapeeloutofavehiclewindow?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF


QCOMP1b. anapplecore,orabananapeelonthegroundwhenyouarewalkingtoorfromyourvehicleortransitarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP1c. anapplecore,orabananapeelonthegroundwhenyouareataparkorotheroutdoorarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP2a. wrappersfromgum,snackfood,orcandyoutofavehiclewindow?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP2b. wrappersfromgum,snackfood,orcandyonthegroundwhenyouarewalkingtoorfromyourvehicleortransitarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP2c. wrappersfromgum,snackfood,orcandyonthegroundwhenyouareataparkorotheroutdoorarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP3a. paperorplasticfastfoodcontainersoutofavehiclewindow?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP3b. paperorplasticfastfoodcontainersonthegroundwhenyouarewalkingtoorfromyourvehicleortransitarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP3c. paperorplasticfastfoodcontainersonthegroundwhenyouareataparkorotheroutdoorarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP4a. gumoutofavehiclewindow?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP4b. gumonthegroundwhenyouarewalkingtoorfromyourvehicleortransitarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP4c. gumonthegroundwhenyouareataparkorotheroutdoorarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF


FOR SMOKERS ONLY [ASK IF QTU3 = 1 OR 2, ELSE SKIP TO TPN]:

QCOMP5a. cigarettebuttsoutofavehiclewindow?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP5b. cigarettebuttsonthegroundwhenyouarewalkingtoorfromyourvehicleortransitarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QCOMP5c. cigarettebuttsonthegroundwhenyouareataparkorotheroutdoorarea?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

PERSONALNORMSSCALE

TPN.Thankyouforyourpatience,wearemorethanhalf‐waydone.NextIamgoingtoaskyouasetoffivesituationalquestions.Therearenorightorwronganswers.Pleaseanswerusingascalefromzerototenwherezeroequalsnopersonalobligationandtenequalsaverystrongpersonalobligation.

Doyoufeelapersonalobligationto...


QPN1. notlitterwhenyouareholdinganemptysoftdrinkcanandtherearenotrashcansavailable?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QPN2. notlitterwhenyouareholdingagumwrapperandtherearenotrashcansavailable?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF


QPN3. stopandpick‐upapieceofscrappaperthatblowsoffabigstackofpapersthatyouarecarryinginbotharms?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QPN4. notlitterwhenyouareill(fever,headache,muscleache)andyouwouldhavetowalkoutofyourwaytoreachatrashreceptacle?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

QPN5. pick‐upapieceofpaperyoudroppedwhenitisrainingandyouaregettingsoaked?

0‐‐‐1‐‐‐2‐‐‐3‐‐‐4‐‐‐5‐‐‐6‐‐‐7‐‐‐8‐‐‐9‐‐‐10‐‐‐98DK‐‐99REF

LITTER BELIEFS AND ACTIONS

TLB.Acknowledgingthatweallhavethepotentialtolitter,I’mgoingtoreadyousixstatementsthatdescribereasonswhyandsituationswhereyoumightlitter.Pleasetellmeifyoustronglyagree,agree,disagree,orstronglydisagreewiththefollowingstatements:

Iammorelikelytolitter...


QLB1. whentheitemI’mholdingisn’trecyclable.

QLB2. whenIaminabadmood.

QLB3. whenIknowsomeoneelsewillbearoundtopickitup.

QLB4. whenIdon’thavetimetotakecareofitanyotherway.

QLB5. whenthereisn’tatrashcanorbagnearby.

QLB6. whentheitemisbiodegradable.

1StronglyAgree‐‐2Agree‐‐3Disagree‐‐4StronglyDisagree‐‐‐‐‐‐‐8DK‐‐9REF


MESSAGES

TM.Now,Ihaveafewquestionsaboutlitterandlitterprevention.

QM1. Using a scale from zero to ten where zero equals not at all important and ten equals very important, in your town or city, how important is the issue of litter?

_________________ 98. DON’T KNOW 99. REFUSED

QM2.Inthepastyear,haveyouseenorheardanymessagesaboutlitterprevention?

1. YES 2. NO [►SKIP TO TD] 8. DON’T KNOW [►SKIP TO TD] 9. REFUSED [►SKIP TO TD]

QM3.Whatwasthesubjectoftheinformationyousaworheard?

[CHECKALLTHATAPPLY]

1. INFORMATION ABOUT GENERAL LITTER 2. INFORMATION ABOUT CIGARETTE BUTT LITTER 3. INFORMATION ABOUT A COMMUNITY/BEACH CLEAN-UP (EVENT) 4. OTHER __________ 8. DON’T KNOW 9. REFUSED

QM4.Wheredidyouseeorhearit?

[CHECKALLTHATAPPLY]

1. PSA ON THE RADIO 2. PSA/COMMERCIAL/NEWS REPORT ON TELEVISION 3. BILLBOARD 4. NEWSPAPER 5. COMMUNITY FLYER/BULLETIN 6. FROM A FRIEND/FAMILY MEMBER 7. OTHER __________ 8. DON’T KNOW 9. REFUSED


DEMOGRAPHICVARIABLES

TD.Wearealmostdone.Thislastsetofquestionsisforclassificationpurposesonly.

QD1.Inwhatyearwereyouborn?


QD2.AreyouofHispanicorLatinoorigin?


QD3.Whatisyourrace?

[IFRESPONDENTSAYSANTHINGOTHERTHANRESPONSESLISTED,PUTINOTHER][FOREXAMPLE,“ITALIAN”OR“BLACKANDJAPANESE”]

1.AMERICANINDIANORALASKANATIVE2.ASIAN3.BLACKORAFRICANAMERICAN4.HISPANICORLATINO5.NATIVEHAWAIIANOROTHERPACIFICISLANDER6.WHITE7.OTHER______________8.DON’TKNOW9.REFUSED


QD4.Whatisthehighestgradeoryearofschoolyouhavecompleted?

[CLARIFY,IFNECESSARY]

1.NEVERATTENDEDSCHOOLORONLYATTENDEDKINDERGARTEN2.GRADES1THROUGH8(ELEMENTARYONLY)3.GRADES9THROUGH11(SOMEHIGHSCHOOL)4.GRADE12ORGED(HIGHSCHOOLGRADUATE)5.SOMECOLLEGEORTECHNICALSCHOOL6.COLLEGEGRADUATE(4‐YEARDEGREE)7.GRADUATECOLLEGEDEGREE8.DON’TKNOW9.REFUSED

QD5.Includingyourself,howmanypeopleliveinyourhousehold?


QD6.Inwhattypeofresidencedoyoulive?

1. House, single detached 2. Apartment/Townhouse/Condominium 3. Mobile Home 8. DON’T KNOW 9. REFUSED

QD7.Whattypeofvehicledoyoudrivemostoften?

1. Car 2. Pick-up truck 3. Van/mini-van 4. Sports Utility Vehicle (SUV) 5. I DON’T DRIVE 6. OTHER ________________ 8. DON’T KNOW 9. REFUSED

QD8.Whatisyourpostalzipcode?



QD9.PleasestopmewhenIgettothecategorythatbestdescribesyourannualhouseholdincome.

1.below$40,000,or[�SKIPTOQD9A]

2.$40,000orabove[�SKIPTOQD9B]

8.DON’TKNOW

9.REFUSED

QD9A.

1.$30,000tounder$40,000

2.$20,000tounder$30,000

3.$10,000tounder$20,000

4.under$10,000

8.DON’TKNOW

9.REFUSED

QD9B.

1.$40,000tounder$50,000

2.$50,000tounder$60,000

3.$60,000tounder$70,000

4.$70,000tounder$80,000

5.$80,000tounder$90,000

6.$90,000tounder$100,000

7.$100,000tounder$150,000

8.$150,000ormore

98.DON’TKNOW

99.REFUSED

CLOSE.Thankyouverymuchforyourtimeandparticipation.


INTERVIEWERONLY

IC1.INTERVIEWER:RECORDGENDER

1. MALE 2. FEMALE 8. DON’T KNOW

IC2.HOWWELLDIDTHERESPONDENTUNDERSTANDTHEQUESTIONS?

1. VERY WELL 2. SOMEWHAT WELL 3. NOT WELL

IC3.HOWATTENTIVEWASTHERESPONDENT?

1. VERY ATTENTIVE 2. SOMEWHAT ATTENTIVE 3. NOT AT ALL ATTENTIVE

IC4.HOWCOOPERATIVEWASTHERESPONDENT?

1. VERY COOPERATIVE 2. SOMEWHAT COOPERATIVE 3. NOT AT ALL COOPERATIVE

Littering Behavior in America: 2009 Report

Documents

Transcript of Littering Behavior in America: 2009 Report