Interviewer Effects from a Total Survey Error Perspective

Section I. History and OverviewThe Past, Present, and Future of Research on Interviewer EffectsIntroductionTraining, Managing, and Monitoring InterviewersInterviewer Effects Across Contexts and ModesInterviewers and NonresponseInterviewer Pace and BehaviorsEstimating Interviewer EffectsClosing ThoughtsAcknowledgmentsReferencesThe Legacy of Charles CannellSummary of Cannell’s Key ContributionsDocumenting Errors in Self-ReportsModeling the Survey Response ProcessInventing New Interviewing TechniquesCreating Methods of Interview ObservationEvidence for the Effectiveness of Interviewer Feedback, Instructions, and CommitmentEarly StudiesSubsequent Record Check InvestigationsSummary and CritiqueMore Recent Cannell-Inspired ResearchInterview ObservationInterviewing MethodsConclusionsReferencesSection II. Training InterviewersGeneral Interviewing Techniques: Developing Evidence-based Practices for Standardized InterviewingIntroductionBrief Historical ContextReasons to Revisit GITA Process for Revising GITRevisiting Interviewing TechniquesGeneral GoalsSpecific Gaps in Existing TrainingStructure of Training and Key ConceptsQuestion–Answer SequenceResponse Format (Question Form)Codable and Uncodable AnswersAcknowledgmentsDiscussionReferencesHow to Conduct Effective Interviewer Training: A Meta-Analysis and Systematic ReviewIntroductionConceptual Development of Research QuestionsEffect of Refusal Avoidance Training on Unit Nonresponse RatesEffect of Interviewer Training on Data QualityEffect Size HeterogeneityTraining Features That May Improve Data QualityData and MethodsResultsWhat Is the Effect of Interviewer Training on Data Quality?Moderator Analysis: Which Features Render Interviewer Training Successful?Interviewer Training Duration (Q4 in Table 4.1)Cooperation Rates and Training Methods (Q5 in Table 4.1)Conclusion and DiscussionAcknowledgmentsBibliographySection III. Managing and Monitoring Interviewers and the Survey ProcessExploring the Mind of the Interviewer: Findings from Research with Interviewers to Improve the Survey ProcessIntroductionMotivationResearch TopicsMethodResultsSensitive QuestionsDifficult QuestionsDeciding Whether to ProbeVignettesSensitivity VignettesDifficulty VignettesSummary and ConclusionsRecommendations for Interviewer TrainingRecommendations for Future ResearchReferencesBehavior Change Techniques for Reducing Interviewer Contributions to Total Survey ErrorIntroductionReview of Relevant LiteratureData and MethodsThe Medical Expenditure Panel Survey (MEPS)Two Critical Question Series: The Calendar Series and Provider Probes SeriesBehavior Coding of Audio-RecordingsSupervisor AlertsResultsBehavior Coding and Rapid FeedbackSupervisor Alert SystemDiscussionBehavior Coding and Rapid FeedbackSupervisor Alert SystemConclusionReferencesStatistical Identification of Fraudulent Interviews in Surveys: Improving Interviewer ControlsIntroductionInterviewer Falsification – An OverviewForms of FalsificationFrequency of FalsificationReasons for FalsificationEffects of Falsification on Data QualityNon-Statistical Identification StrategiesRe-contactMonitoringValidation with Administrative DataStatistical Identification StrategiesIdentification of Complete FalsificationsFalsification IndicatorsMultivariate Analysis of Falsification IndicatorsIdentification of Partial FalsificationsIdentification of Duplicate RecordsData and SampleFindingsIdentification of Complete FalsificationsIdentification of Partial FalsificationsIdentification of Duplicates across the InterviewOutlook and DiscussionAcknowledgmentsReferences/ Examining the Utility of Interviewer Observations on the Survey Response ProcessIntroductionMethodsData SourcesPost-Survey Interviewer Observations in the ESSPost-Survey Interviewer Observations in the NSFGDependent Variables in the ESSDependent Variables in the NSFGAnalytic ApproachResultsESS: Latent Class AnalysisESS: Class Comparisons on Dependent VariablesNSFG: Latent Class AnalysisNSFG: Class Comparisons on Dependent VariablesDiscussionAcknowledgmentsReferencesWhy Do Interviewers Vary in Achieving Interview Privacy and Does Privacy Matter?Introduction and BackgroundMethodsMeasuresAnalysisResultsDiscussionReferencesUnintended Interviewer Bias in a Community-Based Participatory Research Randomized Control Trial among American Indian YouthIntroductionSocial DesirabilityResearch QuestionsMethodsSampleMeasuresAnalytic StrategyResultsInternalizing and Externalizing BehaviorsSubstance Use, Cultural Participation, and Cultural DiscriminationDiscussionLimitationsConclusionsReferencesVirtual Interviewers, Social Identities, and Survey Measurement ErrorIntroductionVirtual InterviewersVirtual Interviewer Identities and Measurement ErrorResearch DesignAnalytic MethodsResultsConclusionsAcknowledgmentsReferencesDifferences in Interaction Quantity and Conversational Flow in CAPI and CATI InterviewsIntroductionNumber of Turns, Events, and WordsUncertainty Markers and HesitationsQuestion CharacteristicsMethodsResultsNumber of Turns and EentsNumber of WordsInterviewer and Respondent Filled Pauses and Respondent Uncertainty MarkersConclusionAcknowledgmentsReferencesInteracting with Interviewers in Text and Voice Interviews on SmartphonesIntroductionStudyAnalyses of Interview DynamicsDo Interviewer Behaviors Predict Response Quality?DiscussionQuestions and ImplicationsAcknowledgmentsReferencesSection V. Interviewers and NonresponseExplaining Interviewer Effects on Survey Unit Nonresponse: A Cross-Survey AnalysisIntroductionDataMethodsResultsSummary and DiscussionReferencesComparing Two Methods for Managing Telephone Interview CasesIntroductionBackground of the Study and Design of the Case Management ApproachesCase Management SystemsInterviewer TrainingInterviewer Monitoring and Case ReviewResultsNumber of Call Attempts Per CaseVariability in Timing of Call AttemptsNumber of Call Attempts Per Interviewer HourOperational ObservationsDiscussionReferencesInvestigating the Use of Nurse Paradata in Understanding Nonresponse to Biological Data CollectionIntroductionDataMethodsParadata VariablesNurse Performance IndicatorsResultsApproach 1: Response Propensity Models Including Nurse Characteristics and Paradata VariablesApproach 2: Response Propensity Models Including Nurse Performance IndicatorsDiscussionReferencesSection VI. Interview Pace and BehaviorsExploring the Antecedents and Consequences of Interviewer Reading Speed (IRS) at the Question LevelIntroductionBackgroundHypothesesMethodsResultsDiscussion and ConclusionsAcknowledgmentsReferencesResponse Times as an Indicator of Data Quality: Associations with Question, Interviewer, and Respondent Characteristics in a Health Survey of Diverse RespondentsIntroductionResponse Times and Question CharacteristicsResponse Times and Interviewers’ ExperienceData and MethodsMeasuresAnalytic StrategyResultsDiscussionAcknowledgmentsReferencesAccuracy and Utility of Using Paradata to Detect Question-Reading DeviationsIntroductionBackgroundInterviewers’ Behavior and Measurement ErrorData and MethodsDataDependent Variable and Variables for QATT Detection MethodsAnalysisResultsSummaryReferencesWhat Do Interviewers Learn?: Changes in Interview Length and Interviewer Behaviors over the Field PeriodIntroductionHypotheses for Behaviors Affected by Interviewer LearningData and MethodsCreating Behavior MeasuresDependent VariablesPrimary Independent Variable: Within-Survey ExperienceControl VariablesAnalytic StrategyResultsRQ1: What Interviewer Behaviors Change over the Course of the Data Collection Period?RQ2: Do Interviewer Behaviors Account for Changes in Survey Length over the Course of the Data Collection Period?Variance ComponentsConclusionsAcknowledgmentsReferencesSection VII. Estimating Interviewer EffectsModeling Interviewer Effects in the National Health Interview SurveyIntroductionMethodsDataMeasuresData StructureStatistical AnalysesResultsIICs by Question CharacteristicsInterviewer IICs by Interviewer CharacteristicsDiscussionReferencesA Comparison of Different Approaches to Examining Whether Interviewer Effects Tend to Vary Across Different Subgroups of RespondentsIntroductionThe Basic ModelData and Preliminary AnalysisDataA Preliminary AnalysisA Two-Step ProcedureThe Conditional Random Interviewer Effect ModelConclusionReferencesDesigning Studies for Comparing Interviewer Variance in Two Groups of Survey InterviewersIntroductionCritical Study Design ConsiderationsInterpenetrated Sample AssignmentGeographic BalanceRandomized Assignment to GroupsPower AnalysisAnalytic ApproachesDiscussionReferences
Next >