當(dāng)代市場調(diào)研 （原第12版）測試題及答案 ch05

Chapter5 Testbank

Chapter5:SampleDesign

MultipleChoice

Underwhatkindofsamplingproceduremusttheresearcheradherecloselytopreciseselectionprocedurestoensurethattheresultsareprojectabletothepopulationunderstudy?

a. nonprobability

b. non-random

c. probability

d. quota

e. noneofthese

Answer:c

________samplesmustbeselectedinsuchawaythateveryelementofthepopulationhasaknownandequalchanceofbeingselected.

a. Cluster

b. Simplerandom

c. Convenience

d. Systematicrandom

e. Stratifiedrandom

Answer:b

_____________isthetotalgroupofpeoplefromwhominformationisobtained.

a. Sampleofinterest

b. Censusofinterest

c. Populationofinterest

d. Subsetofinterest

e. Theentireuniverse

Answer:c

Whatdoconvenience,judgment,quota,andsnowballsampleshaveincommon?

a. Theycanallbeclassifiedasnonprobabilitysamples.

b. Theycanallbeclassifiedasprobabilitysamples.

c. Theyalluserandomnumbertablestogeneratesamples.

d. Ifproperlyinvoked,theyallprovideprojectableresults.

e. Theyutilizeeverynthelementinthesampleframe.

Answer:a

Whichofthefollowingisacorrectstatementregardingprobabilitysamples?

a. Theresearchercanbesureofobtaininginformationfromarepresentativecross-sectionofthepopulationofinterest

b. Samplingerrorcanbecomputed.

c. Thesurveyresultscanbeprojectedtothetotalpopulation.

d. Allofthesestatementsarecorrect.

e. Noneofthesearecorrect.

Answer:d

Whichofthefollowingsamplingmethods,ifproperlyinvoked,willproduceprojectableresults?

a. quota

b. convenience

c. systematic

d. snowball

e. stratifiedquota

Answer:c

Whichofthefollowingtypesofsamplesenablestheresearchertocomputesamplingerror?

a. nonprobability

b. convenience

c. probability

d. allofthese

Answer:c

Whichofthefollowingcouldserveasthebasisforformingastratum?

a. usagerate

b. gender

c. incomelevel

d. allofthese

e. noneofthese

Answer:d

Stratifiedrandomsamplesareoftenconfusedwithwhichofthefollowingnonprobabilitysamplingtechniques?

a. quota

b. convenience

c. judgment

d. snowball

e. noneofthese

Answer:a

Whichofthefollowingsamplingtechniqueswouldgivetheresearcherthemostconfidenceinprojectingthefindingstotheoverallpopulation?

a. quota

b. projectablejudgment

c. simplerandom

d. snowball

e. stratifiedsnowball

Answer:c

Inordertocomputeaskipinterval,theresearcherneedstoknowthedesiredsamplesizeandthe__________.

a. census

b. incidencerate

c. stratifiedsample

d. populationsize

e. noneofthese

Answer:d

Iftheresearcherisexpectingdifferentresponsesbetweenmalesandfemalesinastudy,whichtypeofsamplemightbemostuseful?

a. judgment

b. snowball

c. stratifiedrandom

d. systematicrandom

e. noneofthese

Answer:c

Whichofthefollowingsamplingmethodsisthemoststatisticallyefficient(themostrepresentative)?

a. simplerandom

b. systematicrandom

c. stratifiedrandom

d. cluster

e. Allareequallystatisticallyefficient.

Answer:c

Thetypeoferrorwhichresultsfromthefactthatsamplesarenotperfectlyrepresentativeofthepopulationsfromwhichtheyaredrawnisknownas_______.

a. measurementerror

b. systematicerror

c. samplingerror

d. nonsamplingerror

e. selectionerror

Answer:c

Whichofthefollowingsampletypesisoftenusedasasubstituteforsimplerandomsampling?

a. snowball

b. judgment

c. quota

d. systematic

e. noneofthese

Answer:d

Aresearcherrandomlyselected10censustractsfromametropolitanarea.Withineachcensustract,30householdswererandomlyselected.Thesamplingtechniquejustdescribedwouldbea______________.

a. quotasample

b. areacluster

c. stratifiedrandom

d. systematicrandom

e. systematicquota

Answer:b

Allotherthingsequal,largersampleswillhaveless___________thansmallersamples.

a. measurementerror

b. samplingerror

c. timeandcost

d. nonsamplingerror

e. interviewererror

Answer:b

Thedifferencebetweenthesampleresultandtruepopulationvalueisknownas__________.

a. levelofconfidence

b. nonsamplingerror

c. acceptableerror

d. noneofthese

Answer:c

Thefirststepinthedevelopmentofasamplingplanisto:

a. determinesamplesize.

b. selectasamplingmethod.

c. chooseadatacollectionmethod.

d. definethepopulationofinterest.

e. decidewhichstatisticaltechniquesareappropriate.

Answer:d

The_____________isalistofthemembersorelementsfromwhichunitstobesampledareselected.

a. population

b. sampleframe

c. samplesource

d. sample picture

e. universe

Answer:b

Whatisthemostimportantreasonforthegrowingpopularityofnonprobabilitysamples?

a. samplingerrorcanbecomputed

b. representativeofpopulation

c. lowercostthanprobabilitysamples

d. lessnonresponsebiasthanothertypesofsampling

Answer:c

Ifaresearcheroffersbeerdrinkersanextraincentivetorecruittheirfriendswhoalsodrinkbeertoparticipateinthestudy,s/heisusinga(n)________sample.

a. cluster

b. systematic

c. stratified

d. snowball

e. noneofthese

Answer:d

Ofthefollowing,whichisthemostimportantconsiderationwhenchoosingasample?

a. sizeofthesample

b. sizeofthepopulation

c. howwellthesamplerepresentsthepopulation

d. howwellthepopulationrepresentsthesample

Answer:c

Ina___________,dataareobtainedfromorabouteverymemberofthepopulationofinterest,whileina______________,informationisobtainedfromoraboutapartofthepopulationandusedtomakeestimatesaboutvariouscharacteristicsofthetotalpopulation.

a. sample,census

b. sample,study

c. census,sample

d. census,study

Answer:c

Whichofthefollowingisanadvantagetointernetsampling?

a. respondentscompletesurveyattheirconvenience

b. datacollectionisrelativelyinexpensive

c. theinterviewcanbeadministeredundersoftwarecontrol

d. surveycanbecompletedquickly

e. Allofthese

Answer:e

Whichofthefollowingisanonprobabilitysampleinwhichquotas,basedondemographicorclassificationfactorsselectedbytheresearcher,areestablishedforpopulationsubgroups?

a. Quotasample

b. Conveniencesample

c. Stratifiedsample

d. Systematicsample

e. Snowballsamples

Answer:a

Interviewingcostsaredramaticallyreducedunderwhichofthefollowingsampletype?

a. Systematic

b. Simplerandom

c. Stratified

d. Cluster

e. Noneofthese

Answer:d

The“bigness”ofBigDatacanbeseductiveandleadresearcherstonotquestionwhichofthefollowing?

a. representativeness

b. sampleframe

c. purposefulnonrandomness

d. allofthese

Answer:a

Whichofthefollowingisnotapieceofrequiredinformationinordertodeterminesamplesize?

a. estimatedpopulationstandarddeviation

b. estimatedpopulationmedian

c. acceptablelevelofsamplingerror

d. desiredlevelofconfidence

e. Allarenecessarytodeterminesamplesize.

Answer:b

Theideathatalargenumberofsamplemeansorproportionswillapproximateanormaldistribution,regardlessofthedistributionofthepopulationfromwhichtheyweredrawn,isknownas:

a.samplevariance

b. populationvariance

c. centrallimittheorem

d. randomerror

e. simplerandomsample

Answer:c

Acontinuousdistribution,whichisbell-shapedandsymmetricaboutthemean,isknownas:

a. thenormaldistribution.

b. thechi-squaredistribution.

c. theF-distribution.

d. themeandistribution.

e. thebinomialdistribution.

Answer:a

Aresearcherwantstoknowtheaveragenumberoftimespermonthrespondentseatatfastfoodrestaurants.Thestatisticthats/heismostinterestedinwouldbethe_________.

a. variance

b. mean

c. standarddeviation

d. proportion

Answer:b

Whichofthefollowingisnotadeterminantofsamplesizeforprobabilitysamples?

a. budgetavailable

b. researchobjectives

c. ruleofthumb

d. allofthesearedeterminants

Answer:d

Ifweweremakinginferencesfromasinglesample,wewouldexpectthattherewouldbea____percentprobabilitythatthesamplemeanorproportiongeneratedfromoursampleresultswouldbewithintwostandarderrorsofthetruepopulationmean.

a. 68.26%

b. 99.74%

c. 97.5%

d. 95.44%

e. 98.00%

Answer:d

Higherlevelsofconfidencegenerallyrequire:

a. largersamplesizes.

b. greatervariance.

c. smallersamplesizes.

d. largerstandarddeviation.

Answer:a

Whenaresearchcompanyroutinelyusessamplesof300fortheirstudies,theyareusingwhichtypeofplantodeterminesamplesize?

a. the50%rule

b. budgetavailable

c. ruleofthumb

d. numberofsubgroupstobeanalyzed

Answer:c

Howmanyrespondentsshouldbeineachimportantsubgroupofasample,ataminimum?

a. 10

b. 25

c. 50

d. 95

e. 100

Answer:e

Whichofthefollowingstatementsaboutsamplesizeisfalse?

a. Thereisnodirectrelationshipbetweenpopulationsizeandthesizeofthesamplerequiredtoestimatepopulationparameters.

b. Thelargerthesamplevariance,thelargerthesamplesizethatisrequired.

c. Thehigherthelevelofconfidence,thelargerthesamplesizethatisrequired.

d. Thelargerthepopulation,thelargerthesamplesizethatisrequired.

e. Allofthesestatementsarefalse.

Answer:d

Holdingconstantata95%levelofconfidence,ifthestandarddeviationincreases,thenthesamplesizerequiredtorepresentthepopulationinquestionwillbe_______.

a. larger

b. aboutthesame

c. smaller

d. notapplicable

Answer:a

Asthelevelofacceptableerrorincreasesinthesamplemeansformula,thesamplesizerequiredtorepresentaparticularpopulationis:

a. smaller

b. larger

c. aboutthesame

d. noneofthese

Answer:a

Whichofthefollowingisnotrequiredtodeterminesamplesizeusingthesamplemeansformula?

a. thestandarderrorofthemean

b. desiredlevelofconfidence

c. estimateofpopulationstandarddeviation

d. leveloftolerableerror

e. allofthesearerequired.

Answer:a

Whichofthefollowingisnotamethodforestimatingthepopulationstandarddeviation?

a. priorstudyresults

b. conductapilotstudy

c. usesecondarydata

d. usingpopulationsize

Answer:d

______isarangeofvalueswithinwhichthetruepopulationvalueisestimatedtofall.

a. Intervalestimate

b. Pointestimate

c. Populationestimate

d. Sampleestimate

e. Universeestimate

Answer:a

Whichofthefollowingisnotameasureofcentraltendency?

a. variance

b. mean

c. median

d. mode

Answer:a

Standarderrorofthemeanis:

a. ameasureofdispersionofalargepopulation.

b. ameasureofdispersionoflargesample.

c. thestandarddeviationofadistributionofsamplemeans.

d. theerrorinthecomputationofpopulationstandarddeviation.

e. noneofthese

Answer:c

Allofthestatisticalmethodsfordeterminingsamplesizeinthischapterassumewhichofthefollowing?

a. alargepopulation

b. avalidquestionnaire

c. simplerandomsampling

d. stratifiedsampling

e. noneofthese

Answer:c

Whichofthefollowingwouldincreasethenecessarysamplesizeinastudy?

a. smallermeanofthevariablesinquestion

b. lessneedforprecision

c. needtodividesampleintosub-groups

d. largerpopulation

e. smallerpopulation

Answer:c

About68%oftheareaunderastandardnormalcurvecoversone_________.

a. mean

b. sample

c. variance

d. standarddeviation

e. median

Answer:d

Whenaresearcherisnotsureifpeoplewhorespondedtothesurveyaredifferentfromthosewhodidnotrespond,thisbringsintoquestiontheissueof:

a. inappropriatesamplesize.

b. systematicbias.

c. nonresponsebias.

d. refusalbias.

Answer:c

A(n)____________normaldistributionisanormaldistributionwithameanofzeroandastandarddeviationofone.

a. balanced

b. standard

c. stock

d. ordinary

e. outstanding

Answer:b

The_____________includesallthepossibleindividualsorobjectsfromwhomoraboutwhichinformationisneededtomeettheobjectivesoftheresearch.

a. group

b. selection

c. population

d. range?

e. variance

Answer:c

___________samplingerroristheamountofsamplingerrortheresearcheriswillingtoaccept.

a. Given

b. Projected

c. Accepted

d. Allowable

e. Tolerable

Answer:d

Theresultsofasamplecanbeusedtogeneratewhichofthefollowingconcerningthepopulationmean?

a. pointestimate

b. intervalestimate

c. dotestimate

d. pointandintervalestimate

e. intervalanddotestimate

Answer:d

True/False

Inmarketingresearch,itisalwayspreferabletotakeacensusratherthanasampleofthepopulationunderstudy.

Answer:False

Idealcensusconditionsareseldomavailable,evenwhenthepopulationisverysmall.

Answer:True

Probabilitysamplesaremorepopularinmarketingresearchthannonprobabilitysamples.

Answer:False

OneoftheadvantagesofInternetsurveysisthattargetrespondentscancompletethesurveyattheirconvenience.

Answer:True

AshortcomingofInternetsurveysrelativetoothermodesofdatacollectionistheneedtoinputthedataintoadatabaseafterithasbeencollected.

Answer:False

Marketingresearchersmightviewarespondentwhosefamilyworksinmarketingasasecuritythreat.

Answer:True

Nonprobabilitysamplescostlessthanprobabilitysamples.

Answer:True

Ajudgmentsampleenablestheresearchertomakestatisticalinferencesaboutpopulationvalues.

Answer:False

Stratifiedsamplingisatypeofprobabilitysampling.

Answer:True

Simplerandomsamplingisoftenusedasasubstituteforsystematicsampling.

Answer:False

Thesizeofthepopulationisneededinordertocomputeaskipinterval.

Answer:True

Simplerandomsamplesrequireadetailedlistingofthepopulationofinterest.

Answer:True

Conveniencesamplesshouldnotbeusedinmarketingresearch.

Answer:False

Thekeytomakingaccuratepredictionsaboutthecharacteristicsorbehaviorofalargepopulationonthebasisofarelativelysmallsampleliesinthewayinwhichindividualsareselectedforthesample.

Answer:True

Developinganappropriatesamplingframeisoftenoneoftheeasiestthingsthataresearcherfaces.

Answer:False

Thefinalstepinthesamplingprocessisexecutionoftheoperationalsamplingplan.

Answer:True

Therearefourtypesofprobabilitysamplingmethods:simplerandomsampling,systematicsampling,quotasampling,andclustersampling.

Answer:False

Researchersprefersimplerandomsamplestostratifiedsamplesbecauseoftheirpotentialforgreaterstatisticalefficiency.

Answer:False

Fromthestandpointofstatisticalefficiency,clustersamplesaregenerallylessefficientthanothertypesofprobabilitysamples.

Answer:True

IncreasinglyresearchersareturningtomethodologiesthatinvolveblendingsamplebasedoninterviewscollectedbydifferentmeanssuchasSMS,mailandtelephone.

Answer:True

Systematicsamplingincludesacompromiseforexpediencyandhasaverysmallriskofproducinganonrepresentativesample.

Answer:True

Thegreaterthenumberofsubgroupstobeanalyzed,thesmallerthenecessarysamplesize.

Answer:False

Inanormaldistribution,themean,median,andmodeareequal.

Answer:True

Iflargeenoughsamplesaretaken,mosthumanbehaviorcanbeplottedusinganormaldistributioncurve.

Answer:True

Theoretically,thesamplingdistributionofthemeanisalwaysnormallydistributed.

Answer:True

Populationsizeisthemostimportantdeterminantofeffectivesamplesize.

Answer:False

Anormaldistributionhasatotalof6standarddeviations.

Answer:True

Thelevelofacceptableerror(e)inthesampleproportionsformulaisexpressedintermsofapercent.

Answer:True

Somenormaldistributionscannotbetransformedintostandardnormaldistributions.

Answer:False

Arangeofvalueswithwhichthetruepopulationvalueisestimatedtofallisknownasthepointestimate.

Answer:False

Aconfidenceintervalalwaysincludesthetruepopulationvalue.

Answer:False

Budgetisrarelyaconsiderationwhendeterminingsamplesize.

Answer:False

Subgroupsareoftenchosenbasedondemographicsofthesample,suchasgender,income,etc.

Answer:True

Ifaresearchertakesenoughsamples,thedatawillbegintoapproachanormaldistribution.

Answer:True

The“budgetavailable”approachforcestheresearchertoexplorealternativedata-collectionapproachesandtocarefullyconsiderthevalueofinformationinrelationtocost.

Answer:True

Thetotalareaunderanormalcurveisgreaterthanone.

Answer:False

Thefinitepopulationcorrectionfactor(FPC)isanadjustmenttotherequiredsamplesizethatismadeincaseswherethesampleisexpectedtobeequalto5percentormoreofthetotalpopulation.

Answer:True

Thenormaldistributionisbell-shapedandhasonlyonemode.

Answer:True

Asampledistributionisafrequencydistributionofalltheelementsofanindividual(single)population.

Answer:False

OnlineinterviewingandInternetpanels,alongwithsocial-media–drivensampling,havehadnoimpactconcerningfeasiblesamplesizes.

Answer:False

Aparticularnormaldistributionisuniquelydefinedbyitsmeanandstandarddeviation.

Answer:True

Theproportionalpropertyofanormaldistributionprovidesthebasisforthestatisticalinferences.

Answer:True

Theconfidencelevelistheprobabilitythataparticularintervalwillincludetruepopulationvalue.

Answer:True

Regardlessofhowthetargetsamplesizeisdetermined,theresearcherisconfrontedwiththepracticalproblemoffiguringouthowmanysamplingunitswillberequiredtocompletetheassignment.

Answer:True

Inadditiontobeinginterestedinestimatingsamplingmeans,marketingresearchersfrequentlyareinterestedinestimatingproportionsorpercentages.

Answer:True

EssayQuestions

Aresearcherinvokesapilotsampleandfindsthatrespondentsfromhouseholdswithlessthan$50,000annualincomerespondverydifferentlythanrespondentsfromhouseholdswithgreaterthan$50,000annualincome,withregardtothekeysurveyquestions.Inaddition,theresearcher’spreliminaryresultsshowgreatervarianceamongrespondentsintheover$50,000households.Giventhepreceding,whichprobabilitysamplingmethodshouldtheresearcherinvokeandwhy?

Answer:Whenakeydemographicfactorisrelateddirectlytothekeysurveyquestionsandobjectives,andthereisgreatvarianceamongrespondentsacrossthedemographicfactors,astratifiedrandomsampleshouldbeused.Thisisbecausestratifiedrandomsamplesaremorestatisticallyefficientthanothertypesofprobabilitysamplingtechniques;hence,samplingcosts,whichareusuallylinear,canbereduced.

Aresearcherisdevelopingabodyofknowledgeforthedevelopmentofaformalquestionnaire.Partoftheprocessinvolvesconductingfourfocusgroups.Outlineanappropriatesamplingstrategyforgeneratingthefocus-grouprespondents.

Answer:Theappropriatesamplingprocedureforgeneratingfocusgrouprespondentswouldbeaconveniencesample,iftheinformationneededwaswidelyheldacrossthepopulationofinterest.Iftheinformationneededispossessedonlybysomepotentialrespondents,thenascreenedsamplingprocedure,knownasajudgmentsamplewouldbeused.Ifthebehaviorofinterestislowincidence,thenasnowballsamplewouldbeappropriate.

SupposeamarketingresearcherwasdoingasurveyofsnowskiersresidinginSouthLouisiana,forthespecificpurposeofestimatingthefeasibilityofofferingdiscountedsnowskirentalequipment.Theideawasgaugetheneed,interestandpriceexpectationsofthesesnowskiersastheyheadtoColoradoorsomeothersnowskiingdestination.Whattypeofsamplewouldbeneeded?

Answer:ItisplausibletoassumethattherewillalowincidenceofsnowskiersinSouthLouisiana.Insteadofsurveyingatrandom,theresearchershouldresorttoareferralorsnowballsamplingtechnique.

Whenandwhydoessamplingerroroccur?Ifthesamplingerroremanatesfromsometypeofimproperexecutionofthesamplingplanwhatisitcalled?Canitbemeasured?Whatkindofsamplingerrorcanbemeasuredandhowcanitbereduced?

Answer:Samplingerroroccursbecausesamplesarenotperfectlyrepresentativeofthepopulationsfromwhichtheyaredrawn.SamplingerroremanatingfromimproperexecutionofsamplingproceduresisreferredtoasadministrativeerrorandCANNOTbemeasured.However,randomerrorisduetochanceandcannotbeavoided,butcanbereducedbyincreasingsamplesize.Itcannot,however,betotallyeliminated.

Distinguishbetweenprobabilityandnonprobabilitysamples.Whataretheadvantagesanddisadvantagesofeach?

Answer:Aprobabilitysampleensuresthateachelementwithinthepopulationofinteresthasaknownchanceofbeingchosen.Theadvantagesare:1)representativecross-section,2)samplingerrorcanbecomputed,and3)surveyresultsareprojectabletothetotalpopulation.Thedisadvantagesare:1)expensive,2)morecomplexdesign,and3)ittakesmoretimetoexecute.

Nonprobabilitysamplesdonotensurethateachunitwillhaveaknownchanceofbeingchosen.Discretionisusuallyleftuptotheresearcher.Theadvantagesare:1)costless,2)excellentinexploratoryresearch,3)executionisquick,and4)itcanbeadequatelyrepresentative.Thedisadvantagesare1)thesamplingerrorcannotbecomputed,2)itmaynotberepresentative,and3)resultscannotbeprojectedtothetotalpopulation.Nonprobabilitysamplesarepopularbecausetheycanbequicker,cheaper,andeasiertoconduct.Sometimesnonprobabilitysamplesgivequickanswersthatare“goodenough.”

Assumingtheresearchobjectivesofthestudycanbeaccomplished,researcherswouldmuchprefertoestimatetherequiredsamplesizeusingthesampleproportionformularatherthanthesamplemeansformula.Why?

Answer:Thesamplemeansformularequiresthattheresearchermakeanestimateofthepopulationstandarddeviation.Sincemoststudiesare“new”studies,theresearcherwillprobablyhavetoinvestsometimeinestimatingthepopulationstandarddeviation,usuallyviaapilotsample.Thesampleproportionsformularequiresnosuchestimatefromapilotstudy.

Computetherequiredsamplesizegiventherequiredconfidenceinthesampleresultsis90%.Theleveloftolerablesamplingerroris1%andtheestimatedpopulationstandarddeviationis5.

Answer:Therequiredsamplesizegiventheaboverequirements,usingthesamplemeansformula,wouldbe1,353.

Aresearcherisconferringwithaclienttodointerviewsforapart