reading

Cover
Preface
Section I: Introduction
1. 1 The Promise and Challenge of 3MC Research
  1. 1.1 Overview
  2. 1.2 The Promise
  3. 1.3 The Challenge
  4. 1.4 The Current Volume
  5. References
2. 2 Improving Multinational, Multiregional, and Multicultural (3MC) Comparability Using the Total Survey Error (TSE) Paradigm
  1. 2.1 Introduction
  2. 2.2 Concept of Total Survey Error
  3. 2.3 TSE Interactions
  4. 2.4 TSE and Multiple Surveys
  5. 2.5 TSE Comparison Error in Multinational Surveys
  6. 2.6 Components of TSE and Comparison Error
  7. 2.7 Obtaining Functional Equivalence and Similarity in Comparative Surveys
  8. 2.8 Challenges of Multinational Survey Research
  9. 2.9 Language
  10. 2.10 Structure
  11. 2.11 Culture
  12. 2.12 Resources for Developing and Testing Cross‐national Measures
  13. 2.13 Designing and Assessing Scales in Cross‐national Survey Research
  14. 2.14 TSE and the Multilevel, Multisource Approach
  15. 2.15 Documentation
  16. 2.16 Conclusion
  17. References
3. 3 Addressing Equivalence and Bias in Cross‐cultural Survey Research Within a Mixed Methods Framework
  1. 3.1 Introduction
  2. 3.2 Equivalence and Comparability: Supporting Validity of the Intended Interpretations
  3. 3.3 A Comprehensive Approach to Bias Analysis in 3MC Surveys Within a Mixed Methods Research Framework
  4. 3.4 Closing Remarks
  5. References
Section II: Sampling Approaches
1. 4 Innovative Sample Designs Using GIS Technology
  1. 4.1 Introduction
  2. 4.2 Cluster Selection Stage
  3. 4.3 Household Stage
  4. 4.4 Discussion
  5. References
  6. GIS and Remote Sensing Data Resources
2. 5 Within‐household Selection of Respondents
  1. 5.1 Introduction
  2. 5.2 Within‐household Respondent Selection Methods
  3. 5.3 Within‐household Selection Methods in Cross‐national Surveys: The Case of ESS
  4. 5.4 A Few Practical Challenges of Within‐household Sampling in Cross‐national Surveys
  5. 5.5 Summary and Recommendations
  6. References
Section III: Cross‐cultural Questionnaire Design and Testing
1. 6 Overview of Questionnaire Design and Testing
  1. 6.1 Introduction
  2. 6.2 Review of Questionnaire Design and Testing in a Comparative Context
  3. 6.3 Advances in Questionnaire Design and Testing
  4. 6.4 Conclusions
  5. References
2. 7 Sensitive Questions in Comparative Surveys
  1. 7.1 Sensitivity Issues in a Comparative Context
  2. 7.2 The Definition of Sensitivity
  3. 7.3 Approaches to Minimizing the Effect of Sensitivity
  4. 7.4 Measuring Sensitivity in Cross‐national Contexts
  5. 7.5 New Evidence of Cross‐national Sensitivity: SQS Project
  6. 7.6 Understanding Sensitivity
  7. 7.7 Summary
  8. References
3. 8 Implementing a Multinational Study of Questionnaire Design
  1. 8.1 Introduction
  2. 8.2 Scope of the MSQD
  3. 8.3 Design of the MSQD
  4. 8.4 Experiments Implemented in the MSQD
  5. 8.5 Translation Requirements and Procedures
  6. 8.6 Findings on Feasibility and Limitations Due to Translations and Required Adaptations
  7. 8.7 Example Results
  8. 8.8 Conclusion
  9. Acknowledgments
  10. References
4. 9 Using Anchoring Vignettes to Correct for Differential Response Scale Usage in 3MC Surveys
  1. 9.1 Introduction
  2. 9.2 Reporting Heterogeneity
  3. 9.3 Anchoring Vignettes: Design and Analysis
  4. 9.4 Validity of the Model Assumptions
  5. 9.5 Practical Issues
  6. 9.6 Empirical Demonstration of the Anchoring Vignette Method
  7. 9.7 Sensitivity Analysis: Number of Vignettes and Choices of Vignette Intensity
  8. 9.8 Discussion and Conclusion
  9. References
5. 10 Conducting Cognitive Interviewing Studies to Examine Survey Question Comparability
  1. 10.1 Introduction
  2. 10.2 Cognitive Interviewing as a Study in Validity
  3. 10.3 Conducting a Comparative Cognitive Interviewing Study
  4. 10.4 Real‐World Application
  5. 10.5 Conclusion
  6. References
6. 11 Setting Up the Cognitive Interview Task for Non‐English‐speaking Participants in the United States
  1. 11.1 Introduction
  2. 11.2 Differences in Communication Styles Across Languages and Cultures
  3. 11.3 Implications of Cross‐cultural Differences in Survey Pretesting
  4. 11.4 Setting up the Cognitive Interview Task for Non‐English‐speaking Participants
  5. 11.5 Discussion and Recommendations for Future Studies
  6. Disclaimer
  7. Acknowledgment
  8. References
7. 12 Working Toward Comparable Meaning of Different Language Versions of Survey Instruments
  1. 12.1 Introduction
  2. 12.2 Review of the Literature
  3. 12.3 Motivation for the Current Study: US Census Bureau Spanish Usability Testing
  4. 12.4 The Monolingual and Bilingual Cognitive Testing Study
  5. 12.5 Results of the Cognitive Testing
  6. 12.6 Summary and Conclusions
  7. 12.7 Future Research
  8. Disclaimer
  9. Acknowledgment
  10. References
8. 13 Examining the Comparability of Behavior Coding Across Cultures
  1. 13.1 Introduction
  2. 13.2 Methods
  3. 13.3 Results
  4. 13.4 Discussion
  5. Acknowledgments
  6. References
Section IV: Languages, Translation, and Adaptation
1. 14 How to Choose Interview Language in Different Countries
  1. 14.1 Introduction
  2. 14.2 The Issue of Multilingualism
  3. 14.3 Current Practice of Language Choice in Comparative Surveys
  4. 14.4 Using a Language Survey for Decisions About Language Choice for an Interview: Example of Post‐Soviet Region
  5. 14.5 The Choice of Interview Language on the Level of Individual Respondent
  6. 14.6 Summary
  7. References
2. 15 Can the Language of Survey Administration Influence Respondents’ Answers?
  1. 15.1 Introduction
  2. 15.2 Language, Cognition, and Culture
  3. 15.3 Language of Administration in Surveys of Bilingual Bicultural Respondents
  4. 15.4 Data and Methods
  5. 15.5 Results
  6. 15.6 Discussion and Conclusions
  7. References
3. 16 Documenting the Survey Translation and Monitoring Process
  1. 16.1 Introduction
  2. 16.2 Key Concepts
  3. 16.3 Case Study: The ESENER‐2 Study
  4. 16.4 Translation Documentation from a Project Management Perspective
  5. 16.5 Translation Documentation from the Perspective of Translation Teams
  6. 16.6 Translation Documentation from the Perspective of Applied Translation Research
  7. 16.7 Translation Documentation from the Perspective of Data Analysts
  8. 16.8 Summary and Outlook
  9. References
4. 17 Preventing Differences in Translated Survey Items Using the Survey Quality Predictor
  1. 17.1 Introduction
  2. 17.2 Equivalence in Survey Translation
  3. 17.3 Cross‐cultural Survey Translation and Translation Assessment
  4. 17.4 Formal Characteristics of a Survey Item
  5. 17.5 Using SQP: A Five‐step Procedure for Comparing Item Characteristics Across Languages
  6. 17.6 Questions Evaluated in the ESS Round 5, Round 6, and Round 7
  7. 17.7 Discussion
  8. References
Section V: Mixed Mode and Mixed Methods
1. 18 The Design and Implementation of Mixed‐mode Surveys
  1. 18.1 Introduction
  2. 18.2 Consequences of Mixed‐mode Design
  3. 18.3 Designing for Mixed Mode
  4. 18.4 Auxiliary Data for Assessing and Adjusting Mode Effects
  5. 18.5 Conclusions
  6. Acknowledgment
  7. References
2. 19 Mixed‐mode Surveys
  1. 19.1 Introduction
  2. 19.2 Methods
  3. 19.3 Results
  4. 19.4 Discussion and Conclusions
  5. References
3. 20 Mixed Methods in a Comparative Context
  1. 20.1 Introduction
  2. 20.2 Mixed Methods Data Collection Redefined
  3. 20.3 Considerations about Alternate Sources of Data
  4. 20.4 Examples of Social Science Research Using New Technologies
  5. 20.5 Linking Alternative and Survey Data
  6. 20.6 Mixed Methods with Technologically Collected Data in the 3MC Context
  7. 20.7 Conclusions
  8. Acknowledgments
  9. References
Section VI: Response Styles
1. 21 Cross‐cultural Comparability of Response Patterns of Subjective Probability Questions
  1. 21.1 Introduction
  2. 21.2 State‐of‐art Application of Subjective Probability Questions in Surveys
  3. 21.3 Policy Relevance of Subjective Probability Questions
  4. 21.4 Measurement Mechanism for Subjective Probability Questions
  5. 21.5 Data and Methods
  6. 21.6 Results
  7. 21.7 Discussion
  8. References
2. 22 Response Styles in Cross‐cultural Surveys
  1. 22.1 Introduction
  2. 22.2 Data and Measures
  3. 22.3 OLS Regression Analysis
  4. 22.4 Confirmatory Factor Analysis
  5. 22.5 Latent Class Analysis
  6. 22.6 Multidimensional Unfolding Model
  7. 22.7 Discussion and Conclusion
  8. References
3. 23 Examining Translation and Respondents’ Use of Response Scales in 3MC Surveys
  1. 23.1 Introduction
  2. 23.2 Data and Methods
  3. 23.3 Results
  4. 23.4 Discussion
  5. References
Section VII: Data Collection Challenges and Approaches
1. 24 Data Collection in Cross‐national and International Surveys
  1. 24.1 Introduction
  2. 24.2 Recent Developments in Survey Data Collection
  3. 24.3 Data Collection Challenges Faced in Different Regions of the World
  4. 24.4 Future Directions
  5. References
2. 25 Survey Data Collection in Sub‐Saharan Africa (SSA)
  1. 25.1 Introduction
  2. 25.2 Overview of Common Challenges and Solutions in Data Collection in Sub‐Saharan Africa
  3. 25.3 Strategies and Opportunities
  4. 25.4 Future Developments
  5. References
3. 26 Survey Challenges and Strategies in the Middle East and Arab Gulf Regions
  1. 26.1 Introduction
  2. 26.2 Household and Within‐household Sampling
  3. 26.3 Interviewer–Respondent Gender Matching
  4. 26.4 Nationality‐of‐interviewer Effects
  5. 26.5 Response Scale Heterogeneity
  6. 26.6 Conclusion: Outstanding Challenges and Future Directions
  7. References
4. 27 Data Collection in Cross‐national and International Surveys
  1. 27.1 Introduction
  2. 27.2 Survey Research in the Latin America and Caribbean Region
  3. 27.3 Confronting Challenges with Effective Solutions
  4. 27.4 New Opportunities
  5. 27.5 Conclusion
  6. References
5. 28 Survey Research in India and China
  1. 28.1 Introduction
  2. 28.2 Social Science Surveys in India and China
  3. 28.3 Organizational Structure of Surveys
  4. 28.4 Sampling for Household Surveys
  5. 28.5 Permission and Approvals
  6. 28.6 Linguistic Issues
  7. 28.7 Future Directions: New Modes of Data Collection
  8. References
6. 29 Best Practices for Panel Maintenance and Retention
  1. 29.1 Introduction
  2. 29.2 Retention Rates
  3. 29.3 Panel Maintenance Strategies
  4. 29.4 Study Development and the Harmonization of Field Practices
  5. 29.5 Conclusion
  6. References
7. 30 Collection of Biomeasures in a Cross‐national Setting
  1. 30.1 Introduction
  2. 30.2 Background
  3. 30.3 Types of Biomeasures Collected
  4. 30.4 Logistic Considerations
  5. 30.5 Quality Assurance Procedures
  6. 30.6 Ethical and Legal Issues Across Countries
  7. 30.7 Summary and Conclusions
  8. Acknowledgments
  9. References
8. 31 Multinational Event History Calendar Interviewing
  1. 31.1 Introduction
  2. 31.2 EHC Interviews in a Multinational Setting
  3. 31.3 EHC Interview Administration
  4. 31.4 EHC Interviewer Training
  5. 31.5 Interviewer Monitoring in an International Survey
  6. 31.6 Coding Procedures
  7. 31.7 Evaluation of Interviewer Behavior
  8. 31.8 Feedback Processing Speed
  9. 31.9 Effects of Feedback and Interviewer Effects Across Countries
  10. 31.10 Use of Different Cross‐checks Across Countries
  11. 31.11 Discussion
  12. References
9. 32 Ethical Considerations in the Total Survey Error Context
  1. 32.1 Introduction
  2. 32.2 Ethical Considerations and the TSE Framework
  3. 32.3 Origins and Framework of Human Subjects Protection Standards
  4. 32.4 The Belmont Report and the Components of Human Subjects Protection
  5. 32.5 Final Remarks
  6. Acknowledgment
  7. References
10. 33 Linking Auxiliary Data to Survey Data
  1. 33.1 Introduction
  2. 33.2 Ethical Guidelines and Legal Framework
  3. 33.3 What Constitutes Personal Data?
  4. 33.4 Confidentiality
  5. 33.5 Consent
  6. 33.6 Concluding Remarks
  7. References
Section VIII: Quality Control and Monitoring
1. 34 Organizing and Managing Comparative Surveys
  1. 34.1 Introduction
  2. 34.2 Background
  3. 34.3 Factors That Impact 3MC Survey Organization and Management
  4. 34.4 General Considerations and Survey Quality When Applying Project Management to 3MC Surveys
  5. 34.5 The Application of Project Management to 3MC Surveys
  6. 34.6 Conclusion
  7. References
2. 35 Case Studies on Monitoring Interviewer Behavior in International and Multinational Surveys
  1. 35.1 Introduction
  2. 35.2 Case Studies
  3. 35.3 Conclusion
  4. References
3. 36 New Frontiers in Detecting Data Fabrication
  1. 36.1 Introduction
  2. 36.2 Standard Approaches to Detecting Data Falsification
  3. 36.3 Approaches to Preventing Falsification
  4. 36.4 Additional Challenges
  5. 36.5 New Frontiers in Detecting Fraud
  6. 36.6 A Way Forward
  7. References
Section IX: Nonresponse
1. 37 Comparing Nonresponse and Nonresponse Biases in Multinational, Multiregional, and Multicultural Contexts
  1. 37.1 Introduction
  2. 37.2 Harmonization
  3. 37.3 Data Collection Factors
  4. 37.4 Assessment of Risk of Nonresponse Bias
  5. 37.5 Post‐survey Adjustment
  6. 37.6 Conclusion
  7. References
2. 38 Geographic Correlates of Nonresponse in California
  1. 38.1 Introduction
  2. 38.2 Data and Methods
  3. 38.3 Results
  4. 38.4 Discussion and Limitations
  5. References
3. 39 Additional Languages and Representativeness
  1. 39.1 Introduction
  2. 39.2 Data
  3. 39.3 Methods
  4. 39.4 Results
  5. 39.5 Summary and Conclusion
  6. References
Section X: Multi‐group Analysis
1. 40 Measurement Invariance in International Large‐scale Assessments
  1. 40.1 Introduction
  2. 40.2 Measurement Invariance Review
  3. 40.3 Advances in Measurement Invariance
  4. 40.4 The Stepwise Procedure
  5. 40.5 Evaluation Criteria
  6. 40.6 An Example
  7. 40.7 Conclusion
  8. References
2. 41 Approximate Measurement Invariance
  1. 41.1 Introduction
  2. 41.2 The Multigroup Confirmatory Factor Analysis
  3. 41.3 Illustration
  4. 41.4 Discussion and Conclusion
  5. Acknowledgment
  6. References
Section XI: Harmonization, Data Documentation, and Dissemination
1. 42 Data Harmonization, Data Documentation, and Dissemination
  1. Reference
2. 43 Basic Principles of Survey Data Recycling
  1. 43.1 Introduction
  2. 43.2 The Process of Survey Data Recycling
  3. 43.3 The Logic of SDR
  4. 43.4 Using SDR in Constructing the Harmonized Dataset
  5. 43.5 Conclusions
  6. Acknowledgments
  7. References
3. 44 Survey Data Harmonization and the Quality of Data Documentation in Cross‐national Surveys
  1. 44.1 Introduction
  2. 44.2 Standards for Describing the Survey Process from Sampling to Fieldwork
  3. 44.3 Basis of Quality Assessment in the SDR Project
  4. 44.4 Results
  5. 44.5 Concluding Remarks
  6. References
4. 45 Identification of Processing Errors in Cross‐national Surveys
  1. 45.1 Introduction
  2. 45.2 Data and Methods
  3. 45.3 Results
  4. 45.4 Conclusions
  5. Acknowledgments
  6. References
5. 46 Item Metadata as Controls for Ex Post Harmonization of International Survey Projects
  1. 46.1 Introduction
  2. 46.2 Harmonization Controls and Item Quality Controls
  3. 46.3 The Case for Using Item Metadata
  4. 46.4 Application: Trust in Parliament and Participation in Demonstrations
  5. 46.5 Harmonization Controls
  6. 46.6 On the Impact of Harmonization Controls
  7. 46.7 Item Quality Controls
  8. 46.8 Summary and Conclusions
  9. Acknowledgments
  10. References
6. 47 The Past, Present, and Future of Statistical Weights in International Survey Projects
  1. 47.1 Introduction
  2. 47.2 Weighting as a Procedure of Improving Data Quality
  3. 47.3 Availability of Weights and Weight Types in International Survey Projects
  4. 47.4 Quality of Statistical Weights and Consequences of Errors
  5. 47.5 Comparability of Weights or Weighted Data
  6. 47.6 Summary
  7. Acknowledgments
  8. References
Section XII: Looking Forward
1. 48 Prevailing Issues and the Future of Comparative Surveys
  1. 48.1 Introduction
  2. 48.2 Examples of 3MC Surveys
  3. 48.3 Data Quality and Some Special Features of 3MC Surveys
  4. 48.4 Roger Jowell’s Ten Golden Rules for Cross‐national Studies
  5. 48.5 Quality Management
  6. 48.6 A Changing Survey Landscape
  7. 48.7 Big Data
  8. 48.8 Summary of Prevailing Problems
  9. 48.9 Endnote
  10. References
Wiley Series In Survey Methodology
Index
End User License Agreement

List of Tables

Chapter 02
1. Table 2.1 Typology of surveys by mode and medium.
2. Table 2.2 Categorizing nonresponse error.
Chapter 03
1. Table 3.1 Evaluation of MMR core characteristics in 3MC bias studies.
2. Table 3.2 Approaches to integration for a 3MC mixed methods validation study.
Chapter 04
1. Table 4.1 Summary of sampling approaches.
Chapter 07
1. Table 7.1 Reported sensitivity of different topics in a cross‐national perspective.
Chapter 08
1. Table 8.1 The MSQD implementation across participating organizations.
2. Table 8.2 Overview of the experiments.
3. Table 8.3 Question order experiment on attitudes toward abortion for a married woman.
Chapter 09
1. Table 9.1 Sample distributions by country.
2. Table 9.2 Estimates from ordered probit model and the self‐assessment component of the HOPIT model predicting respondent’s pain level (1 none–5 extreme).
3. Table 9.3 Design of the sensitivity analysis using CHARLS, SHARE, and HRS data.
4. Table 9.4 Estimates from the ordered probit model and the self‐assessment component of the HOPIT models predicting respondents’ pain level (1 = none through 5 = extreme): Comparison results between models with different numbers and choices of vignettes.
Chapter 10
1. Table 10.1 First round summary notes: Is your child too sick to play?
2. Table 10.2 Second round summary notes.
Chapter 11
1. Table 11.1 Frequency of Spanish preinterview interactions: experimental versus conventional interviews.
2. Table 11.2 Initial procedure used in the round 1 interviews.
3. Table 11.3 Enhanced introduction and practice.
Chapter 12
1. Table 12.1 Respondent characteristics.
2. Table 12.2 Concepts monolinguals misunderstood^{^a} more frequently than bilinguals.
3. Table 12.3 Concepts bilinguals misunderstood more frequently than monolinguals.
Chapter 13
1. Table 13.1 Behavior codes employed to identify respondent comprehension and mapping difficulties and interviewer question reading problems.
2. Table 13.2 Problematic and nonproblematic survey questions examined.
3. Table 13.3 Cross‐classified HLM model estimates of respondent, response, and question level characteristics on respondent comprehension difficulty.
4. Table 13.4 Cross‐classified HLM model estimates of respondent, response, and question level characteristics on respondent mapping difficulty.
Chapter 14
1. Table 14.1 Documentation of language choice in different comparative projects.
2. Table 14.2 The choice of interview language in the survey cycle.
3. Table 14.3 Different aspects of multilingualism (% of total population).
4. Table 14.4 Language knowledge of the “native language” (if the first official language in each country – also called “title language” – is named as the native language).
5. Table 14.5 Types of multilingualism in language usage in countries of the former Soviet Union (% from multilingual respondents).
Chapter 15
1. Table 15.1 Relative risk for the English language compliant group versus Spanish across measures with different language effect expectations reflecting stratification by propensity strata (NLAAS, n = 220).
2. Table 15.2 Means for the English and Spanish language compliant groups across measures with different language effect expectations reflecting stratification by propensity strata (NLAAS, n = 220).
3. Table 15.3 Relative risk for the English language group versus Spanish across measures by expectations of language effect (NIS, n = 632).
Chapter 17
1. Table 17.1 Summary of characteristics inventoried by SQP.
2. Table 17.2 Categories for differences in the SQP codes for two languages.
Chapter 19
1. Table 19.1 Frequencies and percentages of persons by data collection mode in the countries that implemented mixed‐mode survey design, ISSP 2011.
2. Table 19.2 Differences in proportion of persons reporting excellent or very good health status, mode effect estimates, and standard errors using the propensity score matching method, ISSP 2011.
3. Table 19.3 Estimated proportion of persons reporting excellent or very good health status and standard errors by response mode, ISSP 2011.
4. Table 19.4 Estimated proportion of persons reporting excellent or very good health status and standard errors using unadjusted data, calibration, and multiple imputation methods, ISSP 2011.
5. Table 19.5 Estimated differences in the proportion of persons reporting excellent or very good health status and standard errors by estimation method, ISSP 2011.
Chapter 21
1. Table 21.1 Path coefficients from structural equation model of item nonresponse and unsure response (item nonresponse and unsure 50 response) of subjective life expectancy question, the 2010 Health and Retirement Study.
Chapter 22
1. Table 22.1 Question wordings, mean, and standard deviation (SD) of the two rating scales, 2012 American National Election Studies.
2. Table 22.2 Estimated logistic regression coefficients and standard errors of race or ethnicity and control variables on extreme response style (ERS) and acquiescent response style (ARS), 2012 American National Election Studies (weighted).
3. Table 22.3 Model fit statistics of confirmatory factor analysis of ARS.
4. Table 22.4 Estimation of Model 4 with two content factors and one ARS factor (weighted).
5. Table 22.5 Estimated regression coefficients and standard errors of race or ethnicity and control variables on acquiescent response style (ARS based on the CFA), 2012 American National Election Studies (weighted).
6. Table 22.6 LCA model fit statistics, 2012 American National Election Studies (weighted).
7. Table 22.7 Estimated regression coefficients and standard errors of ERS and ARS on the Likert scale items, 2012 American National Election Studies (LCA Model 6b) (weighted).
8. Table 22.8 Estimated logistic regression coefficients and standard errors of race or ethnicity and control variables on extreme response style (ERS) and acquiescent response style (ARS) based on the LCA model, 2012 American National Election Studies (weighted).
9. Table 22.9 MUM overall goodness of fit statistics for moral traditionalism and position of Blacks in society using common or group‐specific shifting and scaling threshold parameters, 2012 American National Election Studies.
10. Table 22.10 MUM estimates of parameters and variances of interest for the response style in group‐specific shifting and scaling parameter model, 2012 American National Election Studies.
11. Table 22.11 MUM and CFA mean score estimates, 2012 American National Election Studies.
Chapter 23
1. Table 23.1 Scale labels used in self‐rated health in five studies (and sample sizes).
2. Table 23.2 Logistic regression models predicting likelihood to report “fair.”
Chapter 24
1. Table 24.1 Dimensions of survey context.
Chapter 25
1. Table 25.1 Kenya case study – strategies and opportunities.
Chapter 30
1. Table 30.1 Country‐specific consent rates to biomeasure collections in SHARE Wave 6.^{^a}
Chapter 31
1. Table 31.1 Overview of behavior coding variables.
2. Table 31.2 Overview of data collection by country.
3. Table 31.3 Percentage of audio‐recorded interviews by country.
4. Table 31.4 Feedback processing speed: number of days between interview and VU University feedback.
5. Table 31.5 Feedback processing speed across countries in days.
6. Table 31.6 Interviewer performance across feedback periods per country.
7. Table 31.7 Frequency (absolute and relative) of cueing and cross‐checking variables.
Chapter 34
1. Table 34.1 Aspects of survey production lifecycle typically handled by sponsor.
2. Table 34.2 Aspects of survey production lifecycle typically handled by coordinating center and local teams.
3. Table 34.3 Project management process descriptions by subject groups included in ISO 21500.
Chapter 35
1. Table 35.1 SNMHS quality control indicators by sources of errors.
2. Table 35.2 Example of a summary sheet showing the flag status (1 flagged, 0 not flagged) for each interviewer on single occurrence indicators.
3. Table 35.3 Summary of intra‐interviewer correlations over 48 survey items for 36 countries in six ESS rounds.
4. Table 35.4 Laptop‐level statistics on quality indicators and flagged underperforming interviewers (in bold) for SHARE wave 6.
Chapter 37
1. Table 37.1 Incentives in ESS round 6 (2012/2013).
Chapter 38
1. Table 38.1 Cultural and household resistance constructs mapped to variable sources.
2. Table 38.2 Univariate distributions of model predictors and bivariate associations with nonresponse (unweighted).
3. Table 38.3 Screener interview response predicted by community‐level cultural dimensions and sample unit resistance (unweighted logistic mixed model, n = 884 199).
4. Table 38.4 Adult interview response predicted by community‐level cultural dimensions and sample unit resistance (unweighted logistic mixed model, n = 152 134).
Chapter 39
1. Table 39.1 Coverage, contact, cooperation, and language problems: predicted probabilities (%) from a multivariate logit model using the SHP data in the household recruitment phase.
2. Table 39.2 Coverage, contact rates, cooperation, and language problems: predicted probabilities from a multivariate logit model using the SHP data in the person recruitment phase, given the household participates.
3. Table 39.3 Coverage, contact rates, cooperation, and language problems: predicted probabilities from a multivariate logit model using ESS/MOSAiCH‐ISSP data.
4. Table 39.4 Percentages of population mastering one language of different combinations by sociodemographic characteristics.
5. Table 39.5 Standard deviation of predicted probabilities for sociodemographic characteristics by language proficiency from a multivariate logit model.
Chapter 40
1. Table 40.1 Item wordings, means (variances) on the diagonal, and covariances on the off‐diagonal.
2. Table 40.2 LEGPROT items descriptive statistics – mean (standard deviation) – across 38 countries.
3. Table 40.3 Parameter specifications and model‐data fits for the exact invariance using the traditional approach and unconstrained and constrained invariance using the Bayesian approach.
Chapter 41
1. Table 41.1 RMSEA and CFI differences between the configural, metric, and scalar models.
2. Table 41.2 The influence of prior variance on parameter differences.
3. Table 41.3 Alteration of the intercept values of dataset 1.
Chapter 43
1. Table 43.1 Selected international survey projects.
Chapter 44
1. Table 44.1 Quality indicators by survey project.
2. Table 44.2 Changes in quality indicators between first and last wave for selected survey projects.
3. Table 44.3 Average quality by country/territory^a and time span of survey coverage.
Chapter 45
1. Table 45.1 Description of the survey projects and sources of documentation.
2. Table 45.2 Source variables available per target variable.
3. Table 45.3 Example of illegitimate variable values.
4. Table 45.4 Example of misleading variable values and illegitimate variable values.
5. Table 45.5 Example of contradictory variable values.
6. Table 45.6 Example of variable values discrepancy and lack of variable value labels.
7. Table 45.7 Example of lack of variable value labels.
8. Table 45.8 Distribution of errors and their types per pool of source variables for given target variable.
Chapter 46
1. Table 46.1 Availability of items on trust in parliament and participation in demonstrations in 22 international survey projects.
2. Table 46.2 Comparison of translation of “trust” and “confidence” in languages of the European Social Survey (ESS) and the European Values Survey (EVS).
3. Table 46.3 Diversity of response scales in items about trust in the parliament.
4. Table 46.4 Diversity of response scales in items about trust in parliament.
5. Table 46.5 Harmonization controls pertaining to the phrasing of the source items on participation in demonstrations.
6. Table 46.6 Variation of the time frame in questions about participation in demonstrations.
7. Table 46.7 Effect of harmonization controls on two target variables: trust in parliament and participation in demonstrations.
8. Table 46.8 Comparison of the quality of questions on trust in parliament in questionnaires from Poland, using the Survey Quality Predictor 2.1.
9. Table 46.9 Item nonresponse to questions on trust in the national parliament and participation in demonstrations by international survey projects.
Chapter 47
1. Table 47.1 Percentages of weights containing particular weighting variables in time periods.
2. Table 47.2 Quality of survey weights by survey project.
Chapter 48
1. Table 48.1 Third‐party presence during interviews for the 2011 World Mental Health Survey.
2. Table 48.2 Response rates in PIAAC Cycle 1.

List of Illustrations

Chapter 02
1. Figure 2.1 Total survey error.
2. Figure 2.2 Total survey error: Comparison error.
Chapter 03
1. Figure 3.1 Conceptual framework for 3MC validation studies.
Chapter 04
1. Figure 4.1 Years since last population count.
2. Figure 4.2 Illustration of grid method.
3. Figure 4.3 Illustration of manual cluster creation method.
4. Figure 4.4 Nighttime lights, east coast of South America.
5. Figure 4.5 Satellite photo of residential neighborhood in Mogadishu, Somalia.
6. Figure 4.6 Illustration of Qibla method.
7. Figure 4.7 Photo taken by UAV of possible housing unit [53].
Chapter 05
1. Figure 5.1 Over‐/underrepresentation of females, by type of sample (ESS 1–6).
2. Figure 5.2 Over‐/underrepresentation of females, by type of sample and within‐household selection method (ESS 1–6).
Chapter 07
1. Figure 7.1 Types of sensitive questions and the impact on survey response in cross‐national comparative surveys.
Chapter 09
1. Figure 9.1 Illustration of reporting heterogeneity for cross‐national studies. The horizontal lines with arrows indicate the continuum scales of the domain (pain level). The short vertical lines indicate the cutoff points respondents use to answer the self‐assessment question. The vertical dashed line indicates respondents’ answers to self‐assessment questions. For those whose pain level falls on that line it is an indication they have the same true pain level.
2. Figure 9.2 Comparison of self‐assessments between two respondents from two countries. The horizontal lines with arrows indicate the continuum scales of the domain (pain level). The short vertical lines indicate the cutoff points respondents’ use to answer the self‐assessment question. The vertical dashed line indicates respondents’ answers to self‐assessment questions. Those whose level of true pain falls on that line indicates that they have the same true pain levels. The vertical dash–dot lines indicate respondents’ answers to different vignette questions. V1–V3 represent responses to three vignette questions. SR refers to respondents’ ratings to the self‐assesment question.
3. Figure 9.3 Distribution of self‐rated pain for Sweden, the United States, and China.
Chapter 10
1. Figure 10.1 Visual representation of construct schema.
2. Figure 10.2 Visual representation of construct schema with two interpretations. To eliminate the unintended interpretation of “listening,” the question was reworded as “Does your child have difficulty hearing sounds like people’s voices or music?”
3. Figure 10.3 Visual representation of construct schema with three interpretations.
4. Figure 10.4 Emerging schema from concurrent analysis.
5. Figure 10.5 Q‐Notes project home screen.
6. Figure 10.6 Q‐Notes data entry screen.
7. Figure 10.7 Q‐Notes analysis page.
Chapter 11
1. Figure 11.1 Individualism scores across countries [19].
2. Figure 11.2 Spanish‐speaking respondent discomfort: experimental versus conventional interviews.
Chapter 12
1. Figure 12.1 Self‐reported English‐speaking ability in screener and debriefing interviews.
2. Figure 12.2 Understanding of the foster child concept by English‐speaking ability. Data regarding understanding of this concept were unavailable for two respondents out of 39.
3. Figure 12.3 Understanding of “housemate/roommate” by English‐speaking ability. Data regarding understanding of this concept were unavailable for three respondents out of 39. Again, the number of respondents in our study was very small. We provide this table only for illustrative purposes and want to note that responses from a different small group of cognitive interview respondents might look somewhat different.
Chapter 13
1. Figure 13.1 Comprehension difficulties by race/ethnicity/language and question type.
2. Figure 13.2 Mapping difficulties by race/ethnicity/language and question type.
3. Figure 13.3 Interviewer question reading problems by race/ethnicity/language and question type.
Chapter 16
1. Figure 16.1 The translation process in ESENER‐2, mapped against the documentation classification.
2. Figure 16.2 Translation and adaptation notes (extract).
Chapter 17
1. Figure 17.1 Nonequivalence across linguistic groups.
2. Figure 17.2 Equivalence across linguistic groups.
Chapter 18
1. Figure 18.1 Example carrousel question in Dutch. First question in a series of five (see navigation bar). Previous and Next buttons are disabled (gray). Seven‐point response scale (totally agree – totally disagree).
2. Figure 18.2 Confounding of wanted mode selection and unwanted mode measurement effects in estimating the survey statistic of interest.
Chapter 19
1. Figure 19.1 Percentage of the population that were Internet users for the countries that participated in the Wave 6 World Values Survey.
2. Figure 19.2 Self‐reported health status distribution by survey mode in the ISSP 2011.
Chapter 21
1. Figure 21.1 Structural equation model of response patterns of subjective probability questions (for simplicity, correlation across traits and paths from demographics to personality is included in the model but not presented in the figure).
2. Figure 21.2 Item nonresponse rates and 95% confidence intervals of subjective life expectancy by country, race, ethnicity, and interview language in the 2010 Health and Retirement Study, wave 1 of the English Longitudinal Study of Ageing and wave 1 of the Survey of Health, Ageing and Retirement in Europe.
3. Figure 21.3 Detailed response patterns of subjective life expectancy question by race, ethnicity, and interview language, the 2010 Health and Retirement Study.
Chapter 22
1. Figure 22.1 Latent class analysis model of acquiescent response style (ARS), extreme response style (ERS), and content latent class variables (F1: moral traditionalism, F2: position of Blacks in society), with covariates.
2. Figure 22.2 Histogram of unadjusted confirmatory factor analysis scores of moral traditionalism by groups. (The dashed line represents the group mean score. More positive total scores correspond to more favorable attitudes toward moral traditionalism.)
3. Figure 22.3 Histogram of unadjusted confirmatory factor analysis scores of position of Blacks in today’s society by groups. (The dashed line represents the group mean score. More positive total scores correspond to more favorable attitudes toward position of Blacks in today’s society.)
Chapter 23
1. Figure 23.1 Weighted response distributions of answers to self‐rated health using the unbalanced scale.
2. Figure 23.2 Weighted response distributions of answers to self‐rated health using both unbalanced and balanced scales.
3. Figure 23.3 Weighted mean health scores by SRH responses using unbalanced scale, by surveys.
4. Figure 23.4 Weighted mean health scores by SRH responses using balanced scale, by surveys.
5. Figure 23.5 Predicted probability to report “fair” using the unbalanced scale by health scores and surveys.
6. Figure 23.6 Predicted probability to report “fair” using the balanced scale by health scores and surveys.
Chapter 24
1. Figure 24.1 Worldwide Devex tenders by year.
Chapter 25
1. Figure 25.1 Map of sub‐Saharan Africa statistical capacity by country.
Chapter 29
1. Figure 29.1 Wave‐on‐wave reinterview rates, household panel surveys. BHPS, British Household Panel Survey; HILDA, Household, Income and Labour Dynamics in Australia Survey; PSID, Panel Study of Income Dynamics (US); SHP, Swiss Household Panel; SOEP, Socio‐economic Panel (Germany); UKHLS, UK Household Longitudinal Study. Figures in parentheses indicate the year in which interviewing commenced. The PSID response rates are calculated at the family level, while the rates for all other studies are calculated at the individual level. The rates for the SOEP, BHPS, SHP, and HILDA Survey exclude deaths and moves abroad from the denominator; the rates for the PSID only exclude deaths.
2. Figure 29.2 Reinterview rates, wave 1 respondents. (a) Household panel surveys. (b) Child cohort studies. (c) Youth cohort studies. (d) Older aged cohort studies. BCS, British Cohort Study; BHPS, British Household Panel Survey; CVFPS, Chitwan Valley Family Panel Study (Nepal); ELSA, English Longitudinal Study of Ageing; HILDA, Household, Income and Labour Dynamics in Australia Survey; HRS, Health and Retirement Study (US); IFLS, Indonesian Family Life Survey; KLIPS, Korean Labor and Income Panel Study; LSAC, Longitudinal Survey of Australian Children; LSAY, Longitudinal Study of Australian Youth; LSYPE, Longitudinal Survey of Young People in England; MCS, Millennium Cohort Study (UK); NCDS, National Child Development Study (Great Britain); NLSCY, National Longitudinal Survey of Children and Youth (Canada); NLSY, National Longitudinal Survey of Youth (US); NLYOM, National Longitudinal Surveys of Older Males (US); SHARE, Survey of Health, Ageing and Retirement in Europe; SHP, Swiss Household Panel; SLID, Survey of Labour and Income Dynamics (Canada); SOEP, Socio‐economic Panel (Germany); WLS, Wisconsin Longitudinal Study (US); YCS, Youth Cohort Study (England). Figures in parentheses indicate the year in which interviewing commenced. Generally the response rates are calculated excluding deaths and moves abroad. The rates for the following studies only exclude deaths: IFLS, BCS, WLS, NLSY, YCS, and HRS. The following studies do not exclude either: SLID, LSAC, and SHARE. The IFLS includes an expansion of the eligible respondents in selected households in Year 7. The CVFPS and KLIPS response rates are calculated at the household level (and the CVFPS is averaged across months).
Chapter 31
1. Figure 31.1 Design of the CAPI EHC on laptops.
2. Figure 31.2 Display of behavior coding program.
Chapter 34
1. Figure 34.1 The survey production lifecycle.
Chapter 35
1. Figure 35.1 Head office, verification center, audit team, field branches, and field teams connectivity. HH, household; HO, head office; HRD, human resource development; ORV, offline real‐time verification center.
2. Figure 35.2 Viewership trend for three competitor channels in market A.
3. Figure 35.3 Interviewer‐level random intercepts (EBLUPs) for channel XYZ.
4. Figure 35.4 Example of “question not read” report showing interviewers flagged on any single occurrence of “not read” indicator and the drilling capability by date, sample ID, and question field.
5. Figure 35.5 Data flow during ongoing fieldwork of SHARE.
Chapter 37
1. Figure 37.1 Response rates in round 6 of the ESS (2012/2013).
2. Figure 37.2 Response rates and average number of calls per sample unit in the ESS round 6 (2012/2013).
3. Figure 37.3 Response rate and the use of respondent incentives in the ESS round 6 (2012/2013).
4. Figure 37.4 Survey administration in the ESS round 6 (2012/2013).
Chapter 38
1. Figure 38.1 Cultural Ecosystems Nonresponse (CENR) model.
Chapter 39
1. Figure 39.1 Representation of households and persons by nationality due to reasons for nonobservation.
Chapter 40
1. Figure 40.1 Illustrations of (a) uniform measurement noninvariance and (b) nonuniform measurement noninvariance.
2. Figure 40.2 Differences between groups in the parameter estimates (e.g. differences in the item intercept, ∆τ) using the exact invariance of the ML approach (top) and the Bayesian approach (bottom). The mean of the differences between groups is 0 in both approaches. In the exact invariance, note the group means are exactly the same in that there is no variation in the distribution (i.e. ∆τ = 0 and variance = 0). In the Bayesian approach with approximate invariance, the group means are not exactly the same but approximately so, which means that the average of the differences between groups is 0. This implies some variations between groups (i.e. ∆τ ≈ 0 and variance ≠0). The differences between groups might be large and close to zero, for example, with a variance of 0.10 as in the example in the bottom left plot, or rather small and close to zero, for example, a variance of 0.01 as in the bottom right plot.
3. Figure 40.3 (a) Country means between scores computed from the exact scalar invariance and parsimonious invariance procedures. (b) Correlations between scores computed from the exact scalar invariance and parsimonious invariance procedures.
Chapter 41
1. Figure 41.1 Response functions (lines) for different groups (grayscale) under exact (a) vs. approximate (b) measurement invariance models.
2. Figure 41.2 Mplus input file containing the population parameter values for the intercepts, factor loadings, latent means, and latent variances.
3. Figure 41.3 Mplus output of the MGCFA chi‐square comparisons. The scalar equivalence model fits significantly worse than the metric equivalence model; hence exact measurement equivalence does not hold.
4. Figure 41.4 Input file in Mplus for the Bayesian approximate measurement equivalence test.
5. Figure 41.5 Visualization of the estimation of the intercept y1 in group 1 and group 2.
6. Figure 41.6 Traceplots to judge the convergence of intercept Y1–Y4 in groups 1 and 2. Note that only the last 50 000 (after the gray vertical line) are used for the parameter estimates.
7. Figure 41.7 Part of the Mplus output resulting from the input file in Figure 41.4.
Chapter 43
1. Figure 43.1 General schema of survey data recycling.
Chapter 44
1. Figure 44.1 Changes in survey quality over time.
Chapter 45
1. Figure 45.1 Distribution of processing errors per survey project wave. Note: Survey project waves not mentioned in the figure did not contain processing errors.
2. Figure 45.2 Changes of the data processing quality over time.
Chapter 46
1. Figure 46.1 Transformation of source values into the 0–10 scale of trust in parliament.
Chapter 47
1. Figure 47.1 Percentage of studies using weights per year.
2. Figure 47.2 Types of survey weightings across countries. Darker shade, tendency to use design over post‐stratification weights; lighter shade, tendency to use post‐stratification over design weights. White, no data available.
3. Figure 47.3 Distribution of the percentage of incorrectly calculated weights across countries. White, no data.
4. Figure 47.4 Distribution of standard deviation across countries. Darker shade, higher standard deviation; lighter shade, lower standard deviation; white, no data.
Chapter 48
1. Figure 48.1 Example of process control chart.
2. Figure 48.2 The ESS governance scheme.

WILEY SERIES IN SURVEY METHODOLOGY

Advances in Comparative Survey Methods

Multinational, Multiregional, and Multicultural Contexts (3MC)