市場調(diào)研問卷設計專項培訓

市場調(diào)研問卷設計LogicalStepstoDevelopaGoodQuestionnairePlanwhattomeasure計劃好測量什么Formulatequestionstoobtaintheneededinformation擬定為獲取所需要信息而要提的問題Decideontheorderandwordingofquestionsandthelayoutofthequestionnaire決定問題的順序和措辭，決定問卷布局Usingasample,testthequestionnaireforomissionsandambiguity試訪（剔除太長和含糊的）Correcttheproblems(pretestagain,ifnecessary)最后的修改問卷設計中的普通問題題目順序不合理容易引起被訪者理解上的歧義題目措詞差問卷太長沒有給出完整的選項等QRDesignPlanningWhattoMeasure計劃好測量什么Specifyresearchobjectivesandinformationtobecollected研究目的和要收集的信息RelevanceofthestudyTranslatingResearchObjectivesIntoInformationRequirements將研究目的轉(zhuǎn)換成所需的信息FormattingtheQuestionDecisiontobemaderegardingthedegreeoffreedomtobegiventotherespondentsinansweringthequestionsAlternatives選擇的方法Openendedwithnoclassification沒有任何答復類別的開放題問題Openendedwheretheinterviewerusesprecodedclassificationstorecordtheresponse開放題問題，訪問員可以使用預先編碼分類的答案記錄（半開放題）Closeendedorstructuredformatinwhichaquestionorasupplementarycardpresentstheresponsestobeconsidered封閉式題OpenEndedQuestions

OpenEndedQuestionsAreGoodfortheFollowingCircumstances以下情況適合用開放題Whentherearetoomanyresponsestobelisted,ortheycan‘tbeforeseen答案多，無法羅列，或不可預見Whenverbatimresponsesaredesiredtogivetheflavorofpeople‘sanswersortociteexamples需要逐字逐句記下被訪者的答復以便充分了解人們的回答，或者報告里要引用被訪者的語言OpenEndedQuestions(Contd.)

Advantages優(yōu)點Widerangeofresponses回答廣泛Responsesobtainedwithoutanyinfluence答案不受任何影響Freechoices自由選擇

Disadvantages缺點Articulatenessoftherespondentinpersonalinterview表述帶有連接詞Respondentsmayneglecttostatetheobviousanswers答案不明確Involvessubjectivejudgmentsduringsummarizationandarepronetoerror在總結(jié)時可能會引起主觀判斷錯誤Difficulttotrainandenforceinterviewersuniformityinquestioning,probingandrecording培訓訪問員提問，追問，記錄的難度高Responsesrequirelongertimetoprocessbecauseofcoding編碼耗時Closed-endQuestions ThereAreTwoBasicFormatsforClosedEndedorStructuredQuestions封閉題目的2種基本格式Choicefromalistofresponses從列舉的選項中選擇Appropriatesingle-choiceratingonascale在量表范圍中選擇一個適合的封閉題/半封閉題Closed-endQuestions(Contd.) Advantages優(yōu)點Easiertoanswer容易回答Requirelesseffortbytheinterviewer訪問員操作輕松Tabulationandanalysisiseasier容易處理數(shù)據(jù)和分析Lesspotentialerrorinthewaythequestionisaskedandthewayitisrecorded不容易出錯Theresponsesaredirectlycomparablefromrespondenttorespondent被訪者間的答案可直接比較Closed-endQuestions(Contd.)LimitationsofClosedResponseQuestions缺點Disagreementamongresearchersonthetypeofresponsesthatshouldbelisted不論是否否認同的意意見都要羅羅列出來Theanswertoaclosedresponsequestionwillbereceivednomatterhowrelevantorirrelevantthequestionisinthatcontext無論如如何都要取取得答案Maynotproducemeaningfulresults可能產(chǎn)產(chǎn)生沒有意意義的答案案Closed-endQuestions(Contd.)Limitations缺點Goodquestionsarehardtodevelop好的問題不容易設計Providesfeweropportunitiesofselfexpression沒有太多自我表達的機會Thelistofalternativeresponsesprovidesanswersthatmighthavenotbeenconsideredbytherespondentwhoarereluctanttoadmitignorance,andtherebyselectinga“reasonable”response因此，如果被調(diào)查者既有難以抉擇的心理，也有可能對情況不了解，就應該提供中立答案和“不知道”WORDINGINQUESTIONNAIRECareHastoBeTakenThat人人性性化化考考慮慮Isthevocabularysimple,direct,andfamiliartoallreDoanywordshavevagueorambiguousmeanings?是否有模糊的意思Areanyquestions“double-barreled”?一題多義Areanyquestionsleadingorloaded?誘導或者壓力給被訪者Aretheinstructionspotentiallyconfusing?指引混亂Isthequestionapplicabletoallrespondents?是否適合所有被訪者Arethequestionsofappropriatelength?問題長度適合CHOICEOFWORDSUsecommon/simplewordsthatconsumersarefamiliarwith使使用用消消費費者者熟熟悉悉的的語語言言Complicatedwordscanmakeconsumersfeelstupidand/orleadtomisunderstandingofthequestion.ie.技技術(shù)味味很濃濃的專專業(yè)術(shù)術(shù)語Howmanypiecesofwhiteclothesdidyouwashthelasttimeyoudidthelaundry?Howmanypiecesofcoloredsdidyouwashthelasttimeyoudidthelaundry?So,howmanypiecesofclothesintotaldidyouwashthelasttimeyoudidthelaundry?Neveraskconsumerstodomath千萬不不要讓讓消費費者做做算術(shù)術(shù)題CHOICEOFWORDSTotesttheproducts,Iwouldappreciateitifyou-andonlyyou-wouldusethemformesinceitisonlyyouropinioninwhichIaminterested.Useneutral/unbiasedwords使使用中中性/無歧歧義的的措詞詞Forexample,theword““test”isbiasing.Itcanbeascarywordtorespondents--Thisproductmaynotbesafe.What’’swrongwithit?白老老鼠？？“專家家委員員會建建議道道，……你對對此表表示贊贊成還還是反反對？？”“難道道你不不同意意？””CHOICEOFWORDSWheninterviewingprofessionals,avoidoversimplification訪問問專業(yè)業(yè)人士士時，，不要要太簡簡單口口語化化Beconsistentinterminology保保持術(shù)術(shù)語的的一致致性CHOICEOFWORDSPHRASINGOFQUESTIONS措措詞Howmanypiecesofbubblegumdidyoureatlastyear?時間太長長，容易易忘記Respondentscannotpossiblyrememberpreciselyhowmanypiecesofbubblegumtheyeatlastyear.Theanswertothisquestionislikelytobeveryinaccurate.AvoidIfitistoolonganddrawnout,you’ll“l(fā)ose”therespondent.Additionally,ifthequestioncontainsmultiplechoicealternatives,thelastalternativeismorelikelytobeselectedbytherespondent如果是多項選擇，被訪者可能選擇最后的幾個選項PHRASINGOFQUESTIONS措措詞AtleastonceaweekOnceevery1to2weeksOnceevery2to4weeksOnceevery1to3monthsOnceevery3to6monthsOnceevery6to12monthsOnceayearorlessoftenNeverMultiplechoicequestionsshould:-containallalternatives包括括多個選選項-optionsmustnotoverlap/mustbemutuallyexclusive選選項不能能重疊Mutuallyexclusivechoices:OnceaweekormoreoftenOnceevery2to3weeksOnceamonth/every4weeksOnceevery2to3monthsOnceevery4to6monthsOnceortwiceayearLessoftenthanonceayearNever“Doyouthinkproteinandvitaminaregoodforyourhealth,ornot?”“Dothesalesofferyouatestdrivingandgiveyouenoughtimeonpurchasedecisionmaking?”Double--barreledquestionsshouldberephrasedintoseparatequestions一一題多義應應拆分為兩兩題PHRASINGOFQUESTIONS措詞PHRASINGOFQUESTIONSWouldyounotsayBigbabolisbetterthanTatapiece?Don’tleadthewitness!不要要誘導Clearly,thisquestionsuggeststheanswerorrevealstheinterviewer’sopinion.PHRASINGOFQUESTIONSRespondents’choicesmustalsobecollectivelyexhaustive.選項完整Giverespondentsallpossiblechoices!Collectivelyexhaustivechoices:None1-2times3-4times5-6times7ormoretimesShouldweraisepricestomaintainthequalityofproduce,orshouldwekeepthemaboutthesame?Donotleaveoutcertainalternatives.提問中中不要設一一定的選項項Donotjustifyand/orendorseoneormorealternativesPHRASINGOFQUESTIONSAttributesforratingshould:-beexpressedinthepositive(evennegativeattributes)功能句子子表述應該該是正向的的eg:capiseasytoopen/notdifficulttoopen-notbeexpressedincomparativemode不要設設沒有參照照物的比較較方式eg:thecopyisbetter,priceischeaperPHRASINGOFQUESTIONSPHRASINGOFQUESTIONSOpeningquestionsshouldbeeasyandnon-threatening.開放題容易易和無威脅脅的Establishesrapportwithrespondent.Sensitiveanddifficultquestionsmightresultinrespondentendingtheinterview.CommonprinciplesforALLpointsonALLscalesCompletelistofanswers完完整選項Choicesmustbeunique/notoverlap答案單單一，無重重疊Meaningfuldiscriminations/distinctionsbetweenpoints/withincontext界點間間可判別，，有意義Pointsmustbeunderstood界點可理理解SCALES量表SCALESAnotherwaytolookatscaledifferences:Balanced:平衡Unbalanced:不平平衡PurchaseIntent購買買意向Likingscale喜喜歡程度PriceValueScale物有所值值 Newanddifferentscale與眾不同Balanced:平衡Balancedinwording/words/intent措措詞平衡Distancebetweenpointssame/equal界點等等距Unbalanced:不平平衡Encouragesthosewhowouldnotordinarilychoosetoselectoneofthemildalternatives鼓勵那些不不習慣給出出傾向的人人們作出選選擇SCALESTherearetimeswhenyouwillpurposelyuseanunbalancedscale.Unbalanced:Balanced:ExcellentVerygoodVerygoodGoodGoodFairFairPoorPoorVerypoorUnderstandtheimplicationofoddVs.evennumbersofchoices奇偶數(shù)選選項的含義義Ascalewithanoddnumberofchoicesallowsforaneutraloption.Anevennumberedscaleforces“fencesitters”totakeastand奇數(shù)數(shù)項允許中中立，但偶偶數(shù)項則必必須作出傾傾向選擇Odd:Even:StronglyAgreeStronglyAgreeAgreeAgreeNeutralDisagreeDisagreeStronglydisagreeStronglyDisagreeSCALESWhendesigningscales,considercarefullythenumberofscalepoints.認認真真考考慮慮量量表表的的數(shù)數(shù)量量Ifit’saphonerecall,theshorterthebetter.Longscalesarehardtoreadandharderstillfortherespondenttoremember.電電話話訪訪問問，，越越少少選選項項越越好好Numericscales(suchastheHPT9point--+4to-4scale)canaccommodatemorechoices.數(shù)數(shù)字字量量表表適適合合更更多多選選擇擇Withwordscales,trynottoexceed7choices.文文字字量量表表，，不不要要超超過過7個個選選項項Askyourself--areyoumakingmeaningfuldistinctions?SCALES例1：：請請問問您您在在過過去去的的三三個個月月內(nèi)內(nèi)，，有有沒沒有有去去過過超超市市購購物物呢呢？？有………………………………………………1沒有有…………………………………………2這種種選選項項是是量量表表嗎嗎？？例2：：當當您您選選擇擇冰冰箱箱時時，，請請將將下下列列性性能能按按重重要要性性順順序序填填上上1-6_________經(jīng)經(jīng)久久耐耐用用_________無無噪噪聲聲_________省省電電_________價格格便宜宜_________容容量大大_________名牌牌Q&A，以以下屬屬于哪哪類量量表？？例1：：非常好好/很很好/好/一般般/不不好例2：：完全同同意/有點點同意意/即即不是是同意意也不不是不不同意意/有有點不不同意意/完完全不不同意意例3：：適合10987654321不不適合合Q&A，以以下題題目的的量表表通常常是什什么形形式的的？例：您您的家家庭平平均每每個月月收入入是_______元人人民幣幣。QUESTIONSEQUENCE題題目順順序Sequencemustaddresstheintent/purposeofstudy順順序符符合研研究的的目的的Openwithaneasyandnonthreateningquestion開開始題題目應應該容容易和和不敏敏感的的Thequestionnaireshouldflowsmoothlyandlogicallyfromonetopictothenext題目目間過過渡要要流暢暢Proceedfrombroadgeneralquestionstothemorespecific先廣廣泛問問題，，再到到具體體問題題QUESTIONSEQUENCEExample:StandardProducttestquestioning標標準的的產(chǎn)品品口味味測試試題目目OverallRating(whythefirstquestion?)LikesandDislikesAttributeRatingsDirectQuestionsabouttheproductBackgroundquestionsUsethefunnelapproachwhenorderingquestions---GeneraltoSpecificQUESTIONSEQUENCE購買興趣問題之前提問的問題“非常感興趣”的比例(%)1.沒有問任何問題2.82.只問了產(chǎn)品優(yōu)點16.73.只問了產(chǎn)品的缺點0.04.優(yōu)缺點都問了5.7DataSource:EdwinJ.Gross,““TheEffectofQuestionSequenceonMeasuresofBuyingInterest”QUESTIONSEQUENCEOrderquestionsfrommostimportanttoleastimportant最重要要在先，次次要的在后后Becausesomerespondentswillendtheinterviewbeforecompleting,youwanttobesurethatyouhavegatheredthemostimportantinformation.Demographicquestionsshouldbeattheendoftheinterview,exceptwhenrequiredforscreening.背景資料放放后，除非非是甄別部部分需要Thesearesensitiveforsomerespondentstoanswer.QUESTIONSEQUENCEQUESTIONSEQUENCECompleteallquestionsonatopicbeforemovingtoanewtopic完成一一個主題（（部分）的的所有問題題后才到下下一主題（（部分）QUESTIONSEQUENCEOnProducttests,askdislikesbeforelikes.產(chǎn)產(chǎn)品測試，，先問不喜喜歡，再問問喜歡Thereareseveralreasonsforthis:1)Inminimizingrisk,themostimportantthingistoidentifythenegativesoftheproduct.減減少風險2)Respondentsarealreadypositivelypre-disposed--freeproduct消費者者對免費產(chǎn)產(chǎn)品持肯定定傾向3) Ifyoudonothavesomethingnicetosaydonotsayanythingatall…產(chǎn)產(chǎn)品沒有賣賣點，最好好就不要提提了4)Ifjustsaidallpositivethings,mightbehesitanttosayanythingnegative.不不能全說好好的QUESTIONSEQUENCEInP&G’sProducttest,askdislikefirst,butinConcept/Concept&Usetests,asklikesbeforedislikes.概念測試試中，則次次序相反Why?QUESTIONSEQUENCETopicsshouldbeorderedlogically.Usetransitionalstatementstosmoothoutabruptshiftsinquestioning.使用用過渡句ie.Whenshiftingtodemographics...“Thenextfewquestionsareusedtodividetheinterviewsintogroups...”Beconsciousoforderbias意識到到順序引起起的偏差Answerstoaquestioncanbeaffecteddramaticallybythequestionsthatprecededit.當題答案案可能受前前面題目的的影響明顯顯Considerrotatinglistsofattributes(fromfronttobackorrandomly