Applying Science Towards Understanding Behavior in OrganizationsResearch Issues in OrganizationsExperiments: A ReviewExperiments: Pros and ConsExperimental DesignValiditySamplingSampling TechniquesPost with no Control GroupPre – Post with no Control GroupControl Group with no PretestPre – Post with Control GroupMeasurementUtilizing Individual DifferencesTypes of AssessmentBiographical DataSlide 17InterviewsWork SamplesAssessment CentersDrug TestingLetters of RecommendationPsychological Test CharacteristicsPsychological Test TypesIntegrity TestsPersonality TestsThe Big Five InventoryCognitive TestsSlide 29Slide 30Reliability and ValidityApplying Science Towards Understanding Behavior in OrganizationsChapters 2 & 3Research Issues in OrganizationsApproaches to collecting dataExperimentalObservational/correlationalData collection issuesSamplingHow should we select participants?What impact does it have on the results?Experimental designControlling potential confoundsAssigning participants to experimental conditionsMeasurement issuesDescribing and interpreting the resultsExperiments: A ReviewExperiments - Do changes in one variable (X) “cause” changes in another variable (Y)?Independent Variable (X)condition or event that is manipulated by experimenterDependent Variable (Y)variable that is affected (hopefully) by manipulating independent variableExtraneous Variable(s)any variable other than independent variable that may influence dependent variableExperiments: Pros and ConsAdvantage: Allows conclusions about direct effects of one variable on anotherDisadvantages: Experimental conditions are artificialresults may not “generalize” to the real worldSome questions can’t be tested in an experimentRequire control that is not always available in the “real” worldExperimental DesignControlling potential confoundsGoal of experiment is to “rule out” alternate explanations of what affected dependent variableConfounds are threats to internal validityCan be controlled through appropriate experimental design and proceduresInternal ValidityHistoryMaturationTestingInstrumentationStatistical RegressionSelectionMortalitySelection-MaturationDiffusion of TreatmentExternal ValiditySampleSetting (e.g., culture)Time (e.g., 60s vs. 90s)Replication (lack of)Do the results of this experiment generalize (apply) to settings other than the experimentIs there another reason (other than the independent variable) that could explain the results of the experiment.ValidityHow participants are selected for a study influences the extent to which the results can be applied to a larger group (external validity).A wide variety of techniques are availableTwo Main types of samplingProbabilitypredetermined chance of any individual in the population being selected for the studyNonprobabilityTypically nonrandom samplingSamplingSampling TechniquesProbability Sampling1. Simple random sampling2. Systematic sampling3. Stratified random sampling4. Cluster sampling5. Multistage samplingNonprobability Sampling1. Convenience sampling2. Quota sampling3. Snowball samplingPost with no Control GroupTraining PosttestPre – Post with no Control GroupPretest Training PosttestControl Group with no PretestExperimentalGroupTraining PosttestControl Group Placebo PosttestGroupDifferencesPre – Post with Control GroupPretestExperimentalTrainingPosttestPretest Control PosttestGroupDifferencesGroupDifferencesMeasurementMeasurement – the process of assigning numbers to objects or events according to rules (Linn & Gronlund, 1995).Psychological Measurement – concerned with evaluating individual differences in psychological traits.Trait – descriptive label applied to a group of behaviors (e.g., friendly; intelligent)Utilizing Individual DifferencesPsychologists assume that most traits are normally distributed in the population.e.g., height, intelligence, KSAsPsychologists study:measuring these differencesusing these differences to predict performanceI/O Psychologists typically primarily rely on these as predictors of job performance:Cognitive abilitiesPersonalityTypes of AssessmentBiographical InformationInterviewsWork SamplesLetters of RecommendationPsychological TestsBiographical DataGood questions are about events that are: historicalexternaldiscretecontrollable (by the individual)verifiableequal accessjob relevantnon-invasive(Mael, 1991)Rationale vs. empirical methodBiographical DataStrong criterion validitydrug use, criminal history predicts dysfunctional police behavior (Sarchione et al., 1998)not redundant with personality (McManus & Kelly, 1999)Measurement issuesGeneralizabilityFakingFairnessPrivacy concernsInterviewsStructured vs. UnstructuredInfo. gathering vs. interpersonal behavior sampleSituational interview“How would you handle a circumstance in which you needed the help of a person you did not like?”Measurement issuesstructured has more criterion related validityvalue of unstructured? Illusion of validityGuidelines for structured interviewsWork Samplesperform a task under standardized conditionshistorically were for blue collar jobse.g. use of tools, demonstrate driving skillswhite collar examplesspeech interview for foreign worker, test of basic chemistry knowledge, Measurement issueshigh criterion validity if skills are similar to jobcostly to administerwork best with mechanical, rather than people-oriented tasksAssessment CentersRealistic tasks done in groupsAssessed by multiple of raters rating multiple domainsMultiple methodsin basket group exerciseleaderless group exerciseStrong criterion validity (e.g., teachers, police)overall scores predict job performanceMeasurement issuescostly to administerdifferent ratings on a task too highly correlateddimension ratings not correlated strongly across tasksfix? focus on behavior checklists and rater trainingDrug Testingopinion?People are more accepting of it if job involves risks to others (Paronto, et al., 2002)Measurement issuesreliability is very high, but not perfectValidity?Normands, Salyards, & Mahoney (1990)over 5000 postal service applicantsthose who tested positive had 59% higher absenteeism, 47% more likely to be firedno
View Full Document