市场调查方法(英文版)第九章课件

PowerPoint Presentation by Charlie CookPowerPoint Presentation by Charlie CookThe University of West AlabamaThe University of West AlabamaWilliam G.Zikmund William G.Zikmund Barry J.BabinBarry J.BabinPart 2Part 2Designing Designing Research Research StudiesStudies3 3rdrd Edition Edition市场调查方法(英文版)第九章市场调查方法(英文版)第九章市场调查方法(英文版)第九章课件市场调查方法(英文版)第九章课件Basic Issues in Experimental DesignManipulation of Manipulation of the independent the independent variablevariableSelection and Selection and assignment of assignment of experimental experimental subjectssubjectsControl over Control over extraneous extraneous variablesvariablesSelection and Selection and measurement of measurement of the dependent the dependent variablevariableExperimental Experimental DesignDesign4 2007 Thomson/South-Western.All rights reserved.Basic Issues in Experimental DIssues in Experimental DesignManipulation of the Independent VariableManipulation of the Independent VariableIndependent variable:Independent variable:a variable with values that a variable with values that can be manipulated,or altered,independently of any can be manipulated,or altered,independently of any other variable.other variable.Experimental TreatmentExperimental TreatmentThe term referring to the way an experimental The term referring to the way an experimental variable is manipulated.variable is manipulated.vvExperimental group:Experimental group:a group of subjects to whom an a group of subjects to whom an experimental treatment is administered.experimental treatment is administered.vvControl group:Control group:a group of subjects to whom no experimental a group of subjects to whom no experimental treatment is administered.treatment is administered.5 2007 Thomson/South-Western.All rights reserved.Issues in Experimental DesignMExperimental Design(contd)Selection and Measurement of the Dependent Selection and Measurement of the Dependent VariableVariableDependent variable:Dependent variable:the criterion by which the the criterion by which the results of an experiment are judged;a variable results of an experiment are judged;a variable expected to be dependent on the experimenters expected to be dependent on the experimenters manipulation of the independent variable.manipulation of the independent variable.vvSelecting dependent variables that are relevant and truly Selecting dependent variables that are relevant and truly represent an outcome of interest is crucial.represent an outcome of interest is crucial.vvChoosing the right dependent variable is part of the problem Choosing the right dependent variable is part of the problem definition processdefinition processtthorough problem definition will help the horough problem definition will help the researcher select the most important dependent variable(s)researcher select the most important dependent variable(s)whose results will help managers in decision making.whose results will help managers in decision making.6 2007 Thomson/South-Western.All rights reserved.Experimental Design(contd)SeExperimental Design(contd)Selection and Assignment of Test UnitsSelection and Assignment of Test UnitsTest units:the subjects or entities whose responses to Test units:the subjects or entities whose responses to treatment are measured or observed.treatment are measured or observed.Sample Selection And Random Sampling ErrorsSample Selection And Random Sampling ErrorsSystematic or nonsampling errorSystematic or nonsampling errorvvSubject selection,experimental design,and unrecognized Subject selection,experimental design,and unrecognized extraneous variablesextraneous variablesOvercoming sampling errorsOvercoming sampling errorsvvRandomizationRandomizationvvMatchingMatchingvvRepeated measuresRepeated measuresvvControl over extraneous variablesControl over extraneous variables7 2007 Thomson/South-Western.All rights reserved.Experimental Design(contd)SeExperimental Design(contd)Sampling Errors(contd)Sampling Errors(contd)Experimental ConfoundExperimental ConfoundvvWhen there is an alternative explanation beyond the When there is an alternative explanation beyond the experimental variables for any observed differences in the experimental variables for any observed differences in the dependent variable.dependent variable.vvOnce a potential confound is identified,the validity of the Once a potential confound is identified,the validity of the experiment is severely questioned.experiment is severely questioned.vvSources:Sources:Sampling errorSampling error Systematic errorSystematic error Later-identified extraneous variablesLater-identified extraneous variablesvvCareful experimental design can reduce the likelihood of Careful experimental design can reduce the likelihood of confounds.confounds.8 2007 Thomson/South-Western.All rights reserved.Experimental Design(contd)SaDemand CharacteristicsDemand CharacteristicDemand CharacteristicAn experimental design element or procedure that An experimental design element or procedure that unintentionally provides subjects with hints about the unintentionally provides subjects with hints about the research hypothesis.research hypothesis.Demand EffectDemand EffectOccurs when demand characteristics actually affect Occurs when demand characteristics actually affect the dependent variable.the dependent variable.Experimenter Bias Experimenter Bias The influence of the presence,actions,or comments The influence of the presence,actions,or comments of an experimenter on subjects behavior.of an experimenter on subjects behavior.9 2007 Thomson/South-Western.All rights reserved.Demand CharacteristicsDemand CDemand Characteristics(contd)Reducing Demand CharacteristicsReducing Demand Characteristics1.1.Use an experimental disguise.Use an experimental disguise.2.2.Isolate experimental subjects.Isolate experimental subjects.3.3.Use a“blind”experimental administrator.Use a“blind”experimental administrator.4.4.Administer only one treatment level to each subject.Administer only one treatment level to each subject.10 2007 Thomson/South-Western.All rights reserved.Demand Characteristics(contdEXHIBIT 9.1EXHIBIT 9.1By Smiling or Looking Solemn,Experimenters By Smiling or Looking Solemn,Experimenters Can Modify Subjects BehaviorCan Modify Subjects Behavior11 2007 Thomson/South-Western.All rights reserved.EXHIBIT 9.1By Smiling or LookEstablishing ControlConstancy of ConditionsConstancy of ConditionsSubjects in all experimental groups are exposed to Subjects in all experimental groups are exposed to identical conditions except for the differing identical conditions except for the differing experimental treatments.experimental treatments.CounterbalancingCounterbalancingAttempts to eliminate the confounding effects of order Attempts to eliminate the confounding effects of order of presentation by varying the order of presentation of presentation by varying the order of presentation(exposure)of treatments to subject groups.(exposure)of treatments to subject groups.12 2007 Thomson/South-Western.All rights reserved.Establishing ControlConstancy Ethical Issues in ExperimentationDebriefing experimental subjectsDebriefing experimental subjectsCommunicating the purpose of the experimentCommunicating the purpose of the experimentExplaining the researchers hypotheses about the Explaining the researchers hypotheses about the nature of consumer behavior nature of consumer behavior Attempts to interfere with a competitors test-Attempts to interfere with a competitors test-marketing effortsmarketing effortsSuch acts as changing prices or increasing Such acts as changing prices or increasing advertising to influence(confound)competitors test-advertising to influence(confound)competitors test-marketing results are ethically questionable.marketing results are ethically questionable.13 2007 Thomson/South-Western.All rights reserved.Ethical Issues in ExperimentatFundamental Questions in ExperimentationLaboratory ExperimentLaboratory ExperimentA situation in which the researcher has more A situation in which the researcher has more complete control over the research setting and complete control over the research setting and extraneous variables.extraneous variables.Field ExperimentsField ExperimentsResearch projects involving experimental Research projects involving experimental manipulations that are implemented in a natural manipulations that are implemented in a natural environment.environment.14 2007 Thomson/South-Western.All rights reserved.Fundamental Questions in ExperEXHIBIT 9.EXHIBIT 9.5 5The Artificiality of Laboratory versus Field ExperimentsThe Artificiality of Laboratory versus Field Experiments15 2007 Thomson/South-Western.All rights reserved.EXHIBIT 9.5The Artificiality Issues of Experimental ValidityInternal ValidityInternal ValidityThe extent that an experimental variable is truly The extent that an experimental variable is truly responsible for any variance in the dependent responsible for any variance in the dependent variable.variable.vvDid the experiment answer the question of causal effect?Did the experiment answer the question of causal effect?vvDid the manipulation do what it was supposed(predicted)to Did the manipulation do what it was supposed(predicted)to do?do?Manipulation ChecksManipulation ChecksA validity test of an experimental manipulation to A validity test of an experimental manipulation to make sure that the manipulation does produce make sure that the manipulation does produce differences in the independent variable.differences in the independent variable.16 2007 Thomson/South-Western.All rights reserved.Issues of Experimental ValiditExtraneous Variables Affecting Internal ValidityExtraneous Variables Affecting Internal ValidityInternal Internal ValidityValidityMaturationMaturationTestingTestingInstrumentationInstrumentationSelectionSelectionMortalityMortalityHistoryHistory17 2007 Thomson/South-Western.All rights reserved.Extraneous Variables AffectingEffects of Extraneous Variables on ValidityHistory EffectHistory EffectOccurs when some change other than the Occurs when some change other than the experimental treatment occurs during the course of experimental treatment occurs during the course of an experiment that affects the dependent variable.an experiment that affects the dependent variable.Cohort EffectCohort EffectA change in the dependent variable that occurs A change in the dependent variable that occurs because members of one experimental group because members of one experimental group experienced different historical situations than experienced different historical situations than members of other experimental groups.members of other experimental groups.18 2007 Thomson/South-Western.All rights reserved.Effects of Extraneous VariableEffects of Extraneous Variables(contd)Maturation EffectsMaturation EffectsEffects that are a function of time and the naturally Effects that are a function of time and the naturally occurring events that coincide with growth and occurring events that coincide with growth and experience.experience.Testing EffectsTesting EffectsA nuisance effect occurring when the initial A nuisance effect occurring when the initial measurement or test alerts or primes subjects in a measurement or test alerts or primes subjects in a way that affects their response to the experimental way that affects their response to the experimental treatments.treatments.19 2007 Thomson/South-Western.All rights reserved.Effects of Extraneous VariableEffects of Extraneous Variables(contd)Instrumentation EffectInstrumentation EffectA change in the wording of questions,a change in A change in the wording of questions,a change in interviewers,or a change in other procedures causes interviewers,or a change in other procedures causes a change in the dependent variable.a change in the dependent variable.Selection EffectSelection EffectSample bias that results from differential selection of Sample bias that results from differential selection of respondents for the comparison groups,or a sample respondents for the comparison groups,or a sample selection error.selection error.Mortality Effect(Sample Attrition)Mortality Effect(Sample Attrition)Occurs when some subjects withdraw from the Occurs when some subjects withdraw from the experiment before it is completed.experiment before it is completed.20 2007 Thomson/South-Western.All rights reserved.Effects of Extraneous VariableIssues of Experimental Validity(contd)External ValidityExternal ValidityThe accuracy with which experimental results can be The accuracy with which experimental results can be generalized beyond the experimental subjects.generalized beyond the experimental subjects.vvStudent surrogates:Atypical?Student surrogates:Atypical?Trade-Offs Between Internal and External Trade-Offs Between Internal and External ValidityValidityArtificial laboratory experiments usually are high in Artificial laboratory experiments usually are high in internal validity,while naturalistic field experiments internal validity,while naturalistic field experiments generally have less internal validity,but greater generally have less internal validity,but greater external validity.external validity.21 2007 Thomson/South-Western.All rights reserved.Issues of Experimental ValiditClassification of Experimental DesignsBasic Experimental DesignBasic Experimental DesignAn experimental design in which only one variable is An experimental design in which only one variable is manipulated.manipulated.Diagramming Experimental Designs:SymbolsDiagramming Experimental Designs:Symbols22 2007 Thomson/South-Western.All rights reserved.Classification of ExperimentalExamples of Quasi-Experimental DesignsQuasi-experimental DesignsQuasi-experimental DesignsExperimental designs that do not involve random Experimental designs that do not involve random allocation of subjects to treatment combinations.allocation of subjects to treatment combinations.One Shot Design(After Only):One Shot Design(After Only):X XOO1 1One Group PretestPosttest:One Group PretestPosttest:OO1 1X XOO2 2Static Group Design:Static Group Design:ExperimentalExperimentalX XOO1 1ControlControlOO2 223 2007 Thomson/South-Western.All rights reserved.Examples of Quasi-ExperimentalAlternative Experimental DesignsPretestPretest Posttest Control Group DesignPosttest Control Group Design(Before(BeforeAfter with Control)After with Control)ExperimentalExperimentalR OR O1 1 X O X O2 2ControlControl R OR O3 3 X O X O4 4 Posttest Only Control GroupPosttest Only Control Group(After-Only with Control)After-Only with Control)ExperimentalExperimentalR X OR X O1 1ControlControl R OR O2 2Compromise DesignsCompromise Designs24 2007 Thomson/South-Western.All rights reserved.Alternative Experimental DesigEXHIBIT 9.3EXHIBIT 9.3Product Preference Measure in an ExperimentProduct Preference Measure in an Experiment25 2007 Thomson/South-Western.All rights reserved.EXHIBIT 9.3Product PreferenceTest-MarketingTest MarketingTest MarketingAn experimental procedure that provides an An experimental procedure that provides an opportunity to test a new product or a new marketing opportunity to test a new product or a new marketing plan under realistic market conditions to measure plan under realistic market conditions to measure sales or profit potential.sales or profit potential.Controlled ExperimentationNot just trying Not just trying something outsomething outBut scientific But scientific testingtesting26 2007 Thomson/South-Western.All rights reserved.Test-MarketingTest MarketingCoUses of Test-MarketingIdentifying Product Identifying Product WeaknessesWeaknessesForecasting New Forecasting New Product SuccessProduct SuccessTesting the Testing the Marketing MixMarketing MixTest-Test-MarketingMarketing27 2007 Thomson/South-Western.All rights reserved.Uses of Test-MarketingIdentifyFactors to Consider in Test-market SelectionFactors to Consider in Test-market SelectionTest-market Test-market SelectionSelectionDemographic Demographic Composition and Composition and LifestylesLifestylesCompetitive Competitive SituationSituationMedia Coverage Media Coverage and Efficiencyand EfficiencyMediaMediaIsolationIsolationSelf-Contained Self-Contained Trading AreaTrading AreaOverusedOverusedTest-marketsTest-marketsPopulationPopulationSizeSize28 2007 Thomson/South-Western.All rights reserved.Factors to Consider in Test-maKey Terms and Concepts Independent variableIndependent variable Experimental treatment Experimental treatment Experimental group Experimental group Control groupControl group Dependent variable Dependent variable Test units Test units Systematic(or non-sampling Systematic(or non-sampling error)error)RandomizationRandomization MatchingMatching Repeated measures Repeated measures ConfoundConfound Demand characteristicDemand characteristic Demand effectDemand effect Constancy of conditions Constancy of conditions Counterbalancing Counterbalancing DebriefingDebriefing Laboratory experiment Laboratory experiment TachistoscopeTachistoscope Field experiments Field experiments Internal validity Internal validity Manipulation checkManipulation check History effect History effect Cohort effectCohort effect29 2007 Thomson/South-Western.All rights reserved.Key Terms and ConceptsIndependKey Terms and Concepts(contd)Maturation effects Maturation effects Testing effects Testing effects Instrumentation effectInstrumentation effect Mortality effect(or sample attrition)Mortality effect(or sample attrition)External validity External validity Basic experimental designBasic experimental design Quasi-experimental designsQuasi-experimental designs One-shot designOne-shot design One-group pretestposttest designOne-group pretestposttest design Static group designStatic group design Pretestposttest control group designPretestposttest control group design Posttestonly control group designPosttestonly control group design30 2007 Thomson/South-Western.All rights reserved.Key Terms and Concepts(contd