廣告調(diào)查（第二版）英文 指導(dǎo)手冊 Davis-Instructors-Manual-ch13 Questionnaire Design

AdvertisingResearch:Instructor’sManual

Copyright?2012PearsonEducation,Inc.publishingasPrenticeHall

PAGE

PAGE

AdvertisingResearch:Instructor’sManual

Copyright?2012PearsonEducation,Inc.publishingasPrenticeHall

13.QuestionnaireDesign

ChapterGoals

Afterreadingthischapter,studentsshouldbeableto:

? determinethetypeofquestionnairetouseindifferentresearchsituations

? describeandpreparethepartsofaquestionnaire

? prepareandlayoutthequestionnaire

? explainhowpretestingcandramaticallyimprovethequalityofthequestionnaireandtheinformationthatitcollects

NotestotheInstructor

TheChapterLectureprovidesaguidetokeytopicsandcontent.ThePowerPointslidesarenamed:davis_adresearch_ch13.ppt.

Therearetwoonlinesupplementalreadingsthatcanbeusedtoreinforceconceptsrelatedtoquestionnairedesign.

VERBCampaignEvaluation

Chapter1introducedtheCentersforDiseaseControl’sVERBcampaign,whichhadthebroadgoalofincreasingphysicalactivityofyouthaged9to13(“tweens”).Aswithmostmajoradvertisingcampaigns,periodicevaluationswereconductedtodeterminecampaigneffects.Thischapter’ssupplementalreadingsprovidethequestionnairesusedtocollecttheevaluativedata.Sincethereweretwotargetsforthecampaign,twoquestionnaireswerenecessary:oneforparentsandoneforchildren.

ParentsQuestionnaire

ChapterOnediscussedhowquestionnairecontentisselectedtorespondtodecision-makers’informationneeds.Inthecaseofcampaignevaluationresearch,thequestionnairemusthelpdecision-makersanswerthequestion“Howwelldidthecommunicationcampaignaccomplishexplicitcampaigngoals?”Giventhisperspective,itisnotsurprisingthatthequestionsaskedontheVERBparentsquestionnairereflectthespecificparentalgoalsoftheVERBcampaign

astheyrelatetoparents’attitudes,behaviorsandawarenessoftheVERBcampaignitself.Thisquestionnaireisorganizedasfollows:

QuestionnaireSpecific

Page

A-7:

Questions

Purpose Introductiontotheresearchandinformedconsent

A–8 P1toP2a Confirmationthattheinterviewisbeingconductedwiththe

correctrespondent

A-8

PINTRO

Transitionandfocusonupcomingsetofquestions

A–9

P3,P4

Attitudesandinfluencesregardingchild’sphysical activities

A-9

P_D1,P_D2

Specialquestionsforparentsofchildwithadisability

A-10

P5,P6

Attitudesandinfluencesregardingchild’sphysicalactivities

A-10

P8

Parentreportingofchild’sinvolvementinschoolactivities;

attitudesandinfluencesregardingactivities

A-11,12

P10toP11c

Parentparticipationinphysicalactivities

A-12

P11d

Child’svideogameandmediahabits

A-13

P12toP13d

Parentalinvolvementinchild’sactivities

A-13

P13d

Parentself-participationinphysicalactivities

A-14

P7,P14

Parentalattitudestowardphysicalactivity

A-14

P_14a

Parentreportingofchild’smediahabits

A-15

P_14b,P_14c

Parentalknowledgeofphysicalactivityrecommendations

A-15,16

P_14etoP15

VERBcampaignawareness

A-16to19

P16toP23

Classificationquestions

Childs’Questionnaire

ThequestionsaskedonthechildquestionnairealsoreflectthespecificgoalsoftheVERBcampaignwithregardtochildren’sattitudes,behaviorsandawarenessoftheVERBcampaign.Notehowtheintroductionandquestionshavebeenappropriatelywrittenfortheageandlanguageskillsofyoungerrespondents.Thequestionnaireisorganizedasfollows:

QuestionnaireSpecific

Page Questions Purpose

A-20C1Classificationquestion,usedtoskiptoappropriatequestionsetA-20C2Classificationquestion,usedtoskiptoappropriatequestionsetA-21C3Child’sphysicalactivities(typesinlastsevendays)

A-22C4toC5cAmountoftimechildspentinphysicalactivities(lastsevendays)

A-22C6Classificationquestion,usedtoskiptoappropriatequestionset

A-23C7Child’sphysicalactivities(typesyesterday)

A-24C_7xAmountoftimespentinphysicalactivities(yesterday)

A-24C_7atoC_7cSchool-relatedactivities

A-25,26C8Child’sattitudestowardphysicalandleisureactivities

A-27C_8a,C_8bChild’sself-image

A-27C_8cChild’sviewofothers

A-27C9Child’strialofnewactivities

A-28,29C10toC14Child’sleisureactivities

A-29to34C15toC_23ovVERBcampaignawarenessandunderstanding

GenerationNext

ThePewResearchCenterforthePeople&thePressnotesthat“Anewgenerationhascomeofage,shapedbyanunprecedentedrevolutionintechnologyanddramaticeventsbothathomeandabroad.TheyareGenerationNext,thecohortofyoungadultswhohavegrownupwithpersonalcomputers,cellphonesandtheinternetandarenowtakingtheirplaceinaworldwheretheonlyconstantisrapidchange.”Pewconductedresearchtobetterunderstandtheattitudesandbehaviorsofmembersofthisgeneration(whichconsistsofindividualsbornbetween1981and1988.)

ThetelephonequestionnaireusedfordatacollectionintheGenerationNextresearchstudyisprovidedasanonlinesupplementalreading.(Theoriginalquestionnairewithtoplinedataislocatedat

HYPERLINK/reports/questionnaires/300.pdf

/reports/questionnaires/300.pdf

.Notethattheintroductionandinformedconsentarenotshown.)Asyoureadthroughthequestionnairenotethefollowing:

? theuseofopen-endedquestionswhenevertherewasaneedforunprimedortop-of-mindresponses.

? theuseofboldtypefaceandcapitalletterstodistinguishinterviewerinstructionsfromthesurveyquestions.

?theuseofrandomizationinquestions8,10,12,46,47,48,49,59and60.

Wheneverthereisalonglistofitemsthereisalwaysthepotentialforitems

askedearlyinthelisttoinfluenceorbiasresponsestolateritemsbycreating

animplicitframeofreference.Randomizingtheorderoflistitempresentationacrossrespondentssignificantlyreducesthispotentialformofbias.

? theuseofrotationinquestions4,5,6,9and54.Wheneverarespondentisaskedtochooseoneiteminapair,thepotentialexistsforthefirstitemtoexertmoreinfluencesimplybecauseitwasaskedfirst.Pewattemptedtosignificantlyreducethisformofbiasbyrotatingacrossrespondentstheitemthatwasreadfirst.

ChapterLecture

I.DetermineDataCollectionMethod

Customizequestionnairetodatacollectionmethod.

Slide13-

2

Slide13-

3

Slide13-

4

? Questionnaireswrittenforface-to-faceinterviewslookandreaddifferentlythanquestionnaireswrittenfortelephoneorself-administeredonlinestudies.Absenceofface-to-facecontactgenerallyrequiresalesscomplexquestionnaire.

? Self-andinterviewer-administeredquestionnairesarewrittenforthe

"eye"andcanusepicturesoravarietyofmanipulatives.

? Telephonequestionnairesaretypicallywrittenconversationallyandcannotusepicturesorothervisualaids.

II.DetermineNeedforDisguise

Disguisereferstoextentpurposeofresearchishiddenfromrespondent.

? Undisguisedquestionnairesmakespurposeofresearchexplicittotherespondent,oftenthroughquestionnaire'sintroductionorwordingofspecificquestions.

? Disguisedquestionnairesprovideageneraloverviewoftheresearchbuthideinformationonresearchgoalsandsponsorfromrespondent.

Decisionon“best”approachdependsontheextenttowhichitisbelievedthatrespondent'sknowledgeofpurposeand/orsponsorofresearchwillinfluenceresponses.Greaterthelikelihoodofinfluenceorbias,moreappropriateitbecomestodisguisequestionnaire.

Disguisedandundisguisedquestionnairesprovidedifferentperspectivesonthe

samesetofinformationbecauseeachtypeofquestionnaireprovidesadifferentframeofreferencefortherespondent.Neitherapproachisintrinsicallybetter.Itisjustimportanttokeepaframeofreferenceinmindwhenevaluatingstrengthsandweaknessesofdisguisedversusundisguisedformatforaparticularresearchproblem.

III.CreateQuestionnaireComponents

Slide13-

5

Slide13-

6

Slide13-

7

Slide13-

8

Therearefourmaincomponentsofaquestionnaire:introduction,screener,mainbodyofinformationalquestions,andclassificationquestions.

A.Introduction

Introductionexplainsresearchpurposeandgoalsandasksforrespondent'scooperationandparticipation.Specificwordingisdependenton:(1)needfordisguise,(2)studycomplexity,and(3)sensitivenature(ifany)ofthetypeofdatacollected.

Keyelementsinclude:

? explicitorimplicitreferencetoimportance

? generalinformationonresearchrationaleandgoals

? explicitrequestforparticipation

? reassurancethatthetaskofparticipatingisnotburdensomeortime-consuming

? needfortruthfulanswers

? promiseofconfidentiality

? reassurancethatthisisreal,legitimateresearch

Levelofdetailvariesamongdatacollectionmethods.Introductionsinface-to-faceandtelephoneresearchtendtoberelativelyshort.Theintroductiontoamailorself-administeredquestionnairestendstobelongerandmoredetailed.Inthesesituationsthereisnointerviewerpresenttoanswerrespondent'squestions,especiallywithregardtothestudy'spurposeanduseofthedata.

B.Screener

Screenerworkslikeagate:admitsallindividualswhopossessallofthetargetcharacteristicsandeliminatesallwholackatleastonecharacteristic.

Importantthatthescreeneronlycontainthequestionsnecessarytodistinguishtargetfromnontargetindividuals.Sincethesolepurposeofascreeneristoscreenrespondentsitisimportantbutnonessentialthatthescreenerquestionsshouldbeincorporatedintomainbodyofthequestionnaire.

(HandoutorhavestudentsrefertoFigure13.3intext).Thisisascreenerthatcanbeusedinamall-interceptstudy.Note:importanceofexplicitinstructionsfor(1)followingscreenerflowand(2)determiningwhetherindividualpossessesaudiencecharacteristics.

Slide13-

9

Slide13-

10

Slide13-

11

Slide13-

12

Slide13-

13

Slide13-

14

? Questions1-3screenfordesireddemographicandbehavioralcharacteristics.

? Question4continuesscreeningforbehavioralcharacteristicsandpermitsajumptotheappropriatesetofdiet-relatedquestions.

? Question5screenscurrentdieters’purchaseofreducedcaloriefrozendinners.

? Question6screenspastdieters’purchaseofreducedcaloriefrozendinners.Notethatscreenerisstructuredsothattheonlyindividualswho“pass”theentirescreeneraremembersofthetargetsample.

C.MainBody

Containsquestionsthataddressinformationalneeds.

Questionsmustbesequenced.Keepthefollowingguidelinesinmind:

? Beginwithsimple,nonthreatening,interesting,easytoanswerquestions.Simplequestionseaserespondentintotheinterview.Showthattaskofparticipationisnotdifficult.

? Groupquestionsonthesametopictogether;completeonetopicbeforemovingontoanother.Movingfromtopictotopicwithoutalogicalflowincreasespotentialforshallow,top-of-mindresponses.

? Withinatopicmovefromthemostgeneralquestionstothemostspecific.Oncetherespondentisfocused,itismucheasiertoanswerquestionsofgreaterdetailorspecificity.

? Placedifficultorsensitivequestionsattheendofthequestionnaire.Respondentsmuchmorelikelytheansweronce

theyarecommittedandrapporthasbeenestablished.

Slide13-

15

Slide13-

16

Slide13-

17

Slide13-

18

Slide13-

19

? Avoidbiasingquestionsappearinglaterinthequestionnairewithquestionsaskedearlierinthequestionnaire.Beverycarefulthatquestionsaskedearlyintheinterviewdonotprovidetheinformationrequiredfortheansweringoflaterquestions.

? Addressthemostimportanttopicsfirst.Priortofatiguesettinginandaffectingresponsequality.

D.Classification

Aseriesofquestionsthatcollectrelevantdemographic,brandusage,orbehavioralinformationnotcollectedaspartofscreenerorinmainbodyofquestionnaire.Itprovidesthemeansforexaminingimportantsubgroupswithinalargersample.

IV.PhysicallyPrepareQuestionnaire

Fourareasmustbeaddressed:(1)visualappearance,(2)transitionsbetweenmajortopicareasandbetweenindividualquestions,(3)interviewer/respondentinstructions,and(4)responsecolumncoding.Implementeddifferentlyindifferenttypesofquestionnaires.

A.Self-AdministeredQuestionnaires

Goodvisualappearanceisextremelyimportantinself-administeredquestionnaires.Anyindicationthatthequestionnairewillbetoodifficultor

time-consumingtocompletewillgenerallycausearespondenttorefuseordrop

outofthestudy.Itshouldbeuncluttered,easytofollow,andtypedinaneasytoreadtypeface.

(HandoutorhavestudentsrefertoFigure13.4intext).Thisquestionnairehasgoodvisuallayoutandappeal,itisuncluttered,easytoread,anddemonstratesseveralotheraspectsofgoodpreparationandlayout:

? Questionsaredistinguishedfromresponses.Questionsareshowninboldtype,responsesinnormaltype.

? Responsecodingisunobtrusive.

? Questionsdonotcontinueacrosscolumnsorpages.

? Columnsareusedtohelpmaintaintherespondent'sfocus,save

space,andsimplifyresponse.

B.OnlineQuestionnaires

Slide13-

20

Slide13-

21

Slide13-

22

Slide13-

23

Slide13-

24

Thesehavetheirownuniquesetofconsiderations,whichrelateto(1)formandvisualappearance,(2)rangeofquestiontypesavailableacrosssurveycompanies,and(3)wayinwhichspecificquestionsarepresentedacrosssurveycompanies.

1.QuestionnaireFormandVisualAppearance

Severalconsiderationsregardingquestionnaireformandvisualappearanceneedtobekeptinmindregardlessofwhichonlinesurveycompanyisselected.Theseinclude:

? Theuseofpicturesisnotatrivialdecision.Whenpicturesareused,theyshouldhaveanexplicitresearch-specificfunctionandshouldhavearealandmeaningfullinktoresearchquestion.

? Carefullyconsiderhowmayquestionstopresentonasinglewebpage.

Threeoptions:(1)Placetheentireonasinglewebpage.(2)Use

multiplepagesbutgrouprelatedquestionstogetheronasinglepage.(3)

Useasinglepageforeachquestion.

? Understandtherelationshipbetweenspaceprovidedanddepthofresponseinopen-endedquestions.Respondentsusespaceprovidedasanimplicitvisualcuefortheamountofresponsedesiredbythe

researcherandwillwritemorewhengivenlargerspacesforopen-endedresponses.Carefullymatchtheamountofspaceprovidedwith

anticipatedresponse.

? Notallresponseformatsforclosed-endedquestionsareequivalent.

Considerthisquestion“Whichofthefollowingismostimportantto

youinselectingabreakfastcereal?”wheretherespondentcanchoose

fromelevenpotentialanswers.Optionscanbepresentedasanorderedlist,adropboxthatrequiresscrollingoradropboxthatrequiresnoscrolling.Researchshowsthatscrollingdropboxesshouldbeavoided.Dropboxesareonlyusefulwhentherespondentmustsearchalongalistofresponseoptionsforawell-knowntargetanswer.

2.RangeofQuestionTypes

Almostallonlinesurveycompaniespermituseofthemostcommonquestiontypes,forexample,singleselectionmultiplechoicequestions,checklists,singleandmultipleratingscales,andopen-endedresponses.Beyondthesebasic

questiontypes,however,therearesignificantdifferencesacrosscompanieswithregardtotheirabilitytopresentmorecomplexquestiontypes.Companyselection,therefore,shouldbedonepriortothestartofquestionnairedesignbutafterspecificquestiontypesareknown.

3.AppearanceofSpecificQuestionTypes

Whileallstandardquestiontypesareallowedatalmostallonlinesurveycompanies,theforminwhichquestionsarepresentedvarieswidelyacrosscompanies.

Slide13-

25

Slide13-

26

Slide13-

27

Slide13-

28

Slide13-

29

Slide13-

30

Slide13-

31

Slide13-

32

Slide13-

? PsychDataprovidestwooptionsforpresentationofsingleitemratingscale:columnsandnocolumns.

SurveyMonkeyandZoomerangbothtakedifferentapproachestocolumnformat.

Companiesapproachtopresentationofmulti-itemratingscalesreflectsapproachtosingleitems.Again,therearesignificantdifferencesinapproach,especiallyinhowcolumnlabelsanditemsbeingratedaresetoff.

? Psychdata

? Surveymonkey

? Zoomerang

OnlyPsychDataandZoomerangpermittruerankingquestions.Thesignificantdifference,however,isinhoweachcompanypresentsthisquestiontype.PsychDataallowsrespondentstofillinanumbertoindicatetheiropinionwhileZoomerangdefaultstoamoreconfusingmatrixformat.

C.Transitions

Theseareconversational,connectivematerialthatprovidesasenseofflowandcontinuitytothequestionnaire.Transitionshelprespondentsmaintaintheirfocusduringalongseriesofquestions,givewarningthatachangeintopicorfocusisimminent,andprovidejustificationforsensitiveinformation.

D.InterviewerandRespondentInstructions

Therearetimesininterviewer-administeredsurveyswhenitisnecessarytocommunicateinstructionstotheinterviewer.Instructionsforintervieweraredistinguishedfromthequestionsthemselvesbyusingallcapitalletters,asshowninSlide13-32.

Thesameistrueforskippatterns,asshowninSlide13-33.

33

Slide13-

34

Slide13-

35

Skippatternsinonlinesurveysareprogrammedintothesurvey.Someonlinesurveycompanieswillprintavisualconfirmationofaquestionnaire’sskippattern.Itisalwaysrecommendedthatskipspatternsbecloselyexaminedpriortotheactuallaunchofasurvey.Whendevelopingandpreparingskippatternsmakecertainthat:

? allresponsesareaccountedfor.

? allskipdirectionssendtherespondentsomewhereappropriate.E.ColumnCoding

Statisticalanalysisprogramsneedawayofassociatingthenumericvaluesselectedforaspecificquestionwiththequestionnumber.OnlineandtelephoneCATIsurveysdothisautomatically.Paperandpencilsurveysrequirethatthisbedonemanuallythroughcolumncodingwhichreferstoprocessbywhich

eachuniqueresponseisassignedauniqueplaceintheoverallgroupofdata

collected.ExampleshowninSlide13-34.

Number'32'isacolumncode.Ittellsthecomputerthatnumberappearingincolumn32ofthedatainputisthecodeforrespondents'age(acodedvalueof1through4).

Open-endedquestionsalsoneedtobecoded.

V.InternalEvaluation

Allprojectparticipantsshouldreviewthequestionnairepriortoformalpretestingtomakecertainthat:

? thequestionnaireiscompleteandconcise.Thekeytocompletenessisthatthequestionscontainedinthequestionnaireallcorrespondtoidentifiedinformationalneedsandthatthere

arenoquestionsthatdonotcorrespond.

? questionsthemselvesareclear,unambiguous,andappropriateforthetypeofinformationneeded.

? thelayoutisclearandeasytofollow.

VI.PretesttheQuestionnaire

Pretestisimportantbecauseitoffersinsightsneededforimprovingthequestionnaire'swording,structure,format,andorganization.Problemareas

typicallyofinterestduringapretestinclude:

Slide13-

36

Slide13-

37

Slide13-

38

Slide13-

39

Slide13-

40

Slide13-

? problemswithadministration,Didthequestionnairetakeaslongasexpectedtoadminister?Didittakelonger?Ifitistoolong,whatcanbedonetoshortenthequestionnairewithout

sacrificingimportantquestions?

? problemsrelatedtoquestioncomprehension.Werethereanyquestionsthatwerehardfortherespondenttounderstand?Wasthisdifficultyduetotheamountofinformationcontainedinthequestion?Wasthisdifficultyduetothequestion'slength?Wereanyquestionsambiguous?Howcanproblemquestionsberewrittentomaketheinformationneededexplicitandunambiguous?

? problemsrelatedtoquestiondemands.Weretechnicaltermsusedwithoutdefinition?Howmightthesebeexplainedto

reduceconfusion?Werecommonwordsusedinunusualwaysorcontexts?Doquestionsneedtoberewrittentocollectbetterthe

desiredinformation?

? problemsrelatedtoresponseoptions.Arethereanyquestionsforwhich"Don'tKnow"or"NoAnswer"isthemostcommonresponse?Arethereanyquestionsthat"Other"gathersasignificantpercentageofresponses?Aretheresomescalesthatdonotprovideenoughlevelsofdistinction?

? problemsrelatedtoorganizationandquestionsequencing.Didthequestionnaireflowaswellasexpected?Howwelldidtransitionsandinterviewer/respondentinstructionswork?Doanytransitionsorinstructionsneedtoberewritten?Doanytransitionsneedtobeaddedordeleted?Didallskippatternsworkasexpected?Doanyskippatternsneedstoberevised?Wereanyquestionsaskedlaterinthequestionnairebiasedbyearlierquestions?

A.CognitiveInterviews

Thisentailsadministeringdraftsurveyquestions(toasmallnumberoftargetaudienceindividuals)whilecollectingadditionalverbalinformationaboutthesurveyresponses,whichisusedtodeterminewhetheraquestionisgeneratingrequiredinformation

Twoapproachestocognitiveinterviewingare:thinking-aloudandactive

41

Slide13-

42

Slide13-

43

probing.

1.ThinkAloudMethod

Thismethodinvolvesaskingrespondentstoverbalizetheirthoughtsastheyareansweringasurveyquestion.Theinnterviewerreadseachquestiontotherespondent(oneatatime)andthenrecordsand/ornotestheprocesseachrespondentusedtocomprehendandanswerthequestion.

? Theinterviewerisapassiveparticipantintheprocess,typicallyonlyaskingavariationofthequestion“tellmewhatyouarethinking”aboutthisquestion.

2.ActiveProbing

Theintervieweraskseachsurveyquestiononeatatimeandafterananswerorresponse,theinterviewerprobesfurtherinordertounderstandthebasisoftheresponse.

? Interviewerisanactiveparticipantintheprocess,askingspecificquestionstodevelopabetterunderstandingoftherelationshipbetweendesiredinformationandanswerprovidedbyrespondent.

Mostcommontypesofprobesinclude:

? Comprehension/InterpretationProbe:Whatdoestheterm"socialmedia"meantoyou?

? Paraphrasing:CanyourepeatthequestionIjustaskedinyourownwords?

? Confidencejudgment:Howsureareyouthatyouvisitedasocialnetworkingsiteeleventimesinthepastweek?

? Recallprobe:Howdoyourememberthatyousawadvertisingfor

Twitterontelevision?

? Generalprobes:Howdidyouarriveatthatanswer?Wasthateasyorhardtoanswer?Inoticedthatyouhesitated-tellmewhatyouwerethinking.

Questionnairerevisedinlightofinsightsprovidedbypretesting.

VII.PrepareFieldSupportMaterials

Slide13-

44

Thequestionnaireandappropriatesupportingmaterialsaresenttothefieldservice,interviewsareconductedandthequestionnairesarereturnedfordatacodingandanalysis.Potentialproblemswithfieldservicedatacollectionareminimizedwhenquestionnaireisaccompaniedbytwotypesofsupportmaterials:

? Acoverletterthat(a)identifiesthestudy,(b)itemizesmaterialssenttothefieldservice,(c)discussesdatacollectiontechniques,(d)presentsadetaileddescriptionofthesampleandanyspecialquotas,(e)identifiesanyspecialmaterialorequipmentrequirements,and(f)confirmsstudycost,timing,andproceduresforreturnofthequestionnairesandanyothertest-relatedmaterials.

? Amemodescribingthequestionnaireandanyparticularadministrativedetailsofwhichinterviewersshouldbeaware.

ResponsestoApplicationExercises

1to2.Responseswillvary.

3.Thequestionnaireaskstoomanyquestionsoninfluences.Additionalquestionsregardingattitudestowarddonatingareneeded.Specificquestioncontentwillvary.

Problemswiththequestionnaireinclude:

Columncodesareincorrectforanumberofquestions:Q5,Q9(missing),Q9toQ10

(incorrectsequence),Q11(onlyonecolumnisneeded),Q12(outofsequence),Q13,Q16(out

ofsequence),Q18(missing),Q24(outofsequence),Q25(incorrectcode,outofsequence).

Skippatternsareincorrect.Questionsondonatingbehaviorshouldbesequencedsothatthecorrectquestionsareaskedofdonorsandnon-donors,forexample,Q4isdesignedfornon-donorsandasaresulttheseindividualsshouldnotbeaskedQ5.Beyondtheseproblems,Q6skippatterneliminatespotentialforanyonetobeaskedQ8.Q20skippatternisnotneededasbothoptionsgotothesamequestion.Q10shouldnotbeaskedofthosewhosay“no”toQ9.

Thequestionnaireisredundant,withthesamerequestforinformationbeingaskedmultipletimes,forexample,Q9,Q10,Q15,Q16,Q19.

Wordingandformatformanyquestionsareproblematic:Q8assumes(incorrectly)thatthereisonlyoneresponse,Q11isvagueandopentomultipleinterpretations,Q12andQ14wouldbebetterifbalanced,Q17andQ18wordingismuchtoocomplex,Q21shouldbebrokendownintomultiplenumberedquestionswithappropriateskippattern,Q21needsbetter

balanceonresponseoptions,Q23“howlongago”isvagueastothetimeunitrequested,Q23lastitemisbiased,Q24isnotaquestion,Q28needsbetterorderonoptions.

Revisedquestionnaireswillvary.

4.Responseswillvary.

5.Responseswillvary,butsuggesteditemsforattentionarenotedbelow.

AdBlockingSurvey.Q1shouldincorporateexplicitinterpretationofendpointsandmidpointresponsesinthepull-downoptions,Q3shouldbesplitintotwodifferentquestionswithappropriateskippatternfor“yes”and“no”responses(“no”responsesshouldskiptoQ8),Q4needstousesconsistentterminologywithregardto“reas