[原创]Methodology of Longitudinal Surveys

3848

收藏 2009-03-30

&nbsp; 综列数据，与面板数据稍有不同。2009年新书，只收一元。Methodology of Longitudinal Surveys&nbsp;Contents Preface xv 1 Methods for Longitudinal Surveys 1 Peter Lynn 1.1 Introduction 1 1.2 Types of Longitudinal Surveys 2 1.3 Strengths of Longitudinal Surveys 4 1.3.1 Analysis Advantages 4 1.3.2 Data Collection Advantages 6 1.4 Weaknesses of Longitudinal Surveys 8 1.4.1 Analysis Disadvantages 8 1.4.2 Data Collection Disadvantages 9 1.5 Design Features Specific to Longitudinal Surveys 11 1.5.1 Population, Sampling and Weighting 11 1.5.2 Other Design Issues 13 1.6 Quality in Longitudinal Surveys 15 1.6.1 Coverage Error 15 1.6.2 Sampling Error 16 1.6.3 Nonresponse Error 16 1.6.4 Measurement Error 16 1.7 Conclusions 17 References 18 2 Sample Design for Longitudinal Surveys 21 Paul Smith, Peter Lynn and Dave Elliot 2.1 Introduction 21 2.2 Types of Longitudinal Sample Design 21 2.3 Fundamental Aspects of Sample Design 23 2.3.1 Defining the Longitudinal Population 23 v 2.3.2 Target Variables 24 2.3.3 Sample Size 25 2.3.4 Clustering 26 2.3.5 Treatment of Movers 26 2.3.6 Stratification 27 2.3.7 Variances and Design Effects 27 2.3.8 Selection Probabilities 28 2.4 Other Aspects of Design and Implementation 28 2.4.1 Choice of Rotation Period and Pattern 28 2.4.2 Dealing with Births (and Deaths) 29 2.4.3 Sample Overlap 30 2.4.4 Stability of Units and Hierarchies 31 2.5 Conclusion 32 References 32 3 Ethical Issues in Longitudinal Surveys 35 Carli Lessof 3.1 Introduction 35 3.2 History of Research Ethics 35 3.3 Informed Consent 38 3.3.1 Initial Consent 38 3.3.2 Continuing Consent 39 3.3.3 Consent to Trace Respondents 39 3.3.4 Consent for Unanticipated Activities or Analyses 40 3.3.5 Implications for Consent of Changing Circumstances of Sample Members 40 3.3.6 Consent for Linkage to Administrative Data 41 3.3.7 Using Administrative Data without Full Consent 42 3.3.8 Can Fully Informed Consent be Realised? 43 3.4 Free Choice Regarding Participation 43 3.5 Avoiding Harm 46 3.6 Participant Confidentiality and Data Protection 49 3.6.1 Dependent Interviewing 49 3.6.2 The Treatment of Research Data 50 3.7 Independent Ethical Overview and Participant Involvement 52 Acknowledgements 53 References 53 4 Enhancing Longitudinal Surveys by Linking to Administrative Data 55 Lisa Calderwood and Carli Lessof 4.1 Introduction 55 4.2 Administrative Data as a Research Resource 56 4.3 Record Linkage Methodology 58 4.4 Linking Survey Data with Administrative Data at Individual Level 61 4.4.1 Sampling, Sample Maintenance and Sample Evaluation 61 vi CONTENTS 4.4.2 Evaluation Methodology 62 4.4.3 Supplementing and Validating Survey Data 63 4.5 Ethical and Legal Issues 67 4.5.1 Ethical Issues 68 4.5.2 Legal Issues 68 4.5.3 Disclosure Control 68 4.6 Conclusion 69 References 69 5 Tackling Seam Bias Through Questionnaire Design 73 Jeffrey Moore, Nancy Bates, Joanne Pascale and Aniekan Okon 5.1 Introduction 73 5.2 Previous Research on Seam Bias 74 5.3 SIPP and its Dependent Interviewing Procedures 75 5.3.1 SIPP’s Pre-2004 Use of DI 76 5.3.2 Development of New DI Procedures 76 5.3.3 Testing and Refining the New Procedures 78 5.4 Seam Bias Comparison – SIPP 2001 and SIPP 2004 79 5.4.1 Seam Bias Analysis for Programme Participation and Other ‘Spell’ Characteristics 79 5.4.2 Seam Bias Evaluation for Income Amount Transitions 87 5.5 Conclusions and Discussion 89 Acknowledgements 90 References 90 6 Dependent Interviewing: A Framework and Application to Current Research 93 Annette Ja¨ckle 6.1 Introduction 93 6.2 Dependent Interviewing – What and Why? 94 6.2.1 Data Quality 94 6.2.2 Survey Processes 95 6.3 Design Options and their Effects 95 6.3.1 Reactive Dependent Interviewing 96 6.3.2 Proactive Dependent Interviewing 97 6.4 Empirical Evidence 99 6.4.1 Income Sources 100 6.4.2 Current Earnings 104 6.4.3 Current Employment 105 6.4.4 Labour Market Activity Histories 105 6.4.5 School-Based Qualifications 106 6.5 Effects of Dependent Interviewing on Data Quality Across Surveys 107 CONTENTS vii 6.6 Open Issues 109 Acknowledgements 109 References 110 7 Attitudes Over Time: The Psychology of Panel Conditioning 113 Patrick Sturgis, Nick Allum and Ian Brunton-Smith 7.1 Introduction 113 7.2 Panel Conditioning 114 7.3 The Cognitive Stimulus Hypothesis 116 7.4 Data and Measures 117 7.5 Analysis 118 7.6 Discussion 123 References 124 8 Some Consequences of Survey Mode Changes in Longitudinal Surveys 127 Don A. Dillman 8.1 Introduction 127 8.2 Why Change Survey Modes in Longitudinal Surveys? 128 8.3 Why Changing Survey Mode Presents a Problem 130 8.3.1 Changes in Question Structure 130 8.3.2 Effects of Visual vs. Aural Communication Channels 132 8.3.3 Interviewer Presence 135 8.3.4 How Answers to Scalar Questions are Affected by Visual vs. Aural Communication 136 8.4 Conclusions 137 References 137 9 Using Auxiliary Data for Adjustment in Longitudinal Research 141 Dirk Sikkel, Joop Hox and Edith de Leeuw 9.1 Introduction 141 9.2 Missing Data 142 9.3 Calibration 144 9.4 Calibrating Multiple Waves 147 9.5 Differences Between Waves 149 9.6 Single Imputation 150 9.7 Multiple Imputation 151 9.8 Conclusion and Discussion 153 References 155 10 Identifying Factors Affecting Longitudinal Survey Response 157 Nicole Watson and Mark Wooden 10.1 Introduction 157 10.2 Factors Affecting Response and Attrition 159 viii CONTENTS 10.2.1 Locating the Sample Member 159 10.2.2 Contacting the Sample Member 160 10.2.3 Obtaining the Cooperation of the Sample Member 162 10.2.4 The Role of Respondent Characteristics 164 10.3 Predicting Response in the HILDA Survey 167 10.3.1 The HILDA Survey Data 168 10.3.2 Estimation Approach 169 10.3.3 Explanatory Variables 169 10.3.4 Results 171 10.4 Conclusion 178 References 179 11 Keeping in Contact with Mobile Sample Members 183 Mick P. Couper and Mary Beth Ofstedal 11.1 Introduction 183 11.2 The Location Problem in Panel Surveys 184 11.2.1 The Likelihood of Moving 185 11.2.2 The Likelihood of Being Located, Given a Move 187 11.3 Case Study 1: Panel Study of Income Dynamics 190 11.4 Case Study 2: Health and Retirement Study 196 11.5 Discussion 200 Acknowledgements 201 References 202 12 The Use of Respondent Incentives on Longitudinal Surveys 205 Heather Laurie and Peter Lynn 12.1 Introduction 205 12.2 Respondent Incentives on Cross-Sectional Surveys 206 12.2.1 Effects of Incentives on Response Rates on Mail Surveys 206 12.2.2 Effects of Incentives on Response Rates on Interviewer-Administered Surveys 207 12.2.3 Effects of Incentives on Sample Composition and Bias 207 12.2.4 Effects of Incentives on Data Quality 208 12.2.5 Summary: Effects of Incentives 208 12.3 Respondent Incentives on Longitudinal Surveys 208 12.4 Current Practice on Longitudinal Surveys 211 12.5 Experimental Evidence on Longitudinal Surveys 218 12.5.1 Previous Experiments on UK Longitudinal Surveys 221 12.5.2 British Household Panel Survey Incentive Experiment 222 12.6 Conclusion 229 Acknowledgements 230 References 231 CONTENTS ix 13 Attrition in Consumer Panels 235 Robert D. Tortora 13.1 Introduction 235 13.2 The Gallup Poll Panel 237 13.3 Attrition on the Gallup Poll Panel 241 13.3.1 Descriptive Analysis 241 13.3.2 Experiments 244 13.3.3 Logistic Regression 246 13.3.4 A Serendipitous Finding: The Relationship Between Type of Survey and Attrition 246 13.4 Summary 248 References 248 14 Joint Treatment of Nonignorable Dropout and Informative Sampling for Longitudinal Survey Data 251 Abdulhakeem A. H. Eideh and Gad Nathan 14.1 Introduction 251 14.2 Population Model 255 14.3 Sampling Design and Sample Distribution 256 14.3.1 Theorem 1 256 14.3.2 Theorem 2 257 14.4 Sample Distribution Under Informative Sampling and Informative Dropout 257 14.5 Sample Likelihood and Estimation 258 14.5.1 Two-Step Estimation 259 14.5.2 Pseudo Likelihood Approach 259 14.6 Empirical Example: British Labour Force Survey 260 14.7 Conclusions 262 References 263 15 Weighting and Calibration for Household Panels 265 Ulrich Rendtel and Torsten Harms 15.1 Introduction 265 15.2 Follow-up Rules 266 15.2.1 Population Definitions 266 15.2.2 Samples and Follow-up 268 15.3 Design-Based Estimation 269 15.3.1 The Horvitz–Thompson Estimator 269 15.3.2 Link Functions 270 15.3.3 Convexity and Variance of the Weighted Estimator 273 15.4 Calibration 274 15.4.1 Types of Calibration within Panels 275 15.4.2 Bias and Variance 276 15.5 Nonresponse and Attrition 278 x CONTENTS 15.5.1 Empirical Evidence Regarding Nonresponse and Attrition 278 15.5.2 Treatment via Model-Based Prediction 281 15.5.3 Treatment via Estimation of Response Probabilities 281 15.6 Summary 285 References 285 16 Statistical Modelling for Structured Longitudinal Designs 287 Ian Plewis 16.1 Introduction 287 16.2 Methodological Framework 288 16.3 The Data 290 16.4 Modelling One Response from One Cohort 292 16.5 Modelling One Response from More Than One Cohort 295 16.6 Modelling More Than One Response from One Cohort 297 16.7 Modelling Variation Between Generations 298 16.8 Conclusion 300 References 301 17 Using Longitudinal Surveys to Evaluate Interventions 303 Andrea Piesse, David Judkins and Graham Kalton 17.1 Introduction 303 17.2 Interventions, Outcomes and Longitudinal Data 304 17.2.1 Form of the Intervention 304 17.2.2 Types of Effects 305 17.2.3 Conditions for the Evaluation 305 17.2.4 Controlling for Confounders in the Analysis 306 17.2.5 Value of Longitudinal Surveys 307 17.3 Youth Media Campaign Longitudinal Survey 309 17.4 National Survey of Parents and Youth 311 17.5 Gaining Early Awareness and Readiness for Undergraduate Programs (GEAR UP) 314 17.6 Concluding Remarks 315 References 315 18 Robust Likelihood-Based Analysis of Longitudinal Survey Data with Missing Values 317 Roderick Little and Guangyu Zhang 18.1 Introduction 317 18.2 Multiple Imputation for Repeated-Measures Data 318 18.3 Robust MAR Inference with a Single Missing Outcome 320 18.4 Extensions of PSPP to Monotone and General Patterns 323 18.5 Extensions to Inferences Other than Means 324 18.6 Example 325 CONTENTS xi 18.7 Discussion 328 Acknowledgements 330 References 330 19 Assessing the Temporal Association of Events Using Longitudinal Complex Survey Data 333 Norberto Pantoja-Galicia, Mary E. Thompson and Milorad S. Kovacevic 19.1 Introduction 333 19.2 Temporal Order 334 19.2.1 Close Precursor 334 19.2.2 Nonparametric Test for Close Precursor 335 19.3 Nonparametric Density Estimation 335 19.3.1 Kernel Density Estimation 336 19.3.2 Local Likelihood Approach 337 19.4 Survey Weights 340 19.4.1 Assessing the Standard Error 341 19.5 Application: The National Population Health Survey 341 19.5.1 Pregnancy and Smoking Cessation 342 19.5.2 Subsample 342 19.5.3 Interval-Censored Times 342 19.5.4 Results 343 19.6 Application: The Survey of Labour and Income Dynamics 344 19.6.1 Job Loss and Separation or Divorce 345 19.6.2 Subsample 345 19.6.3 Interval-Censored Times 345 19.6.4 Results 346 19.7 Discussion 347 Acknowledgements 348 References 348 20 Using Marginal Mean Models for Data from Longitudinal Surveys with a Complex Design: Some Advances in Methods 351 Georgia Roberts, Qunshu Ren and J.N.K. Rao 20.1 Introduction 351 20.2 Survey-Weighted GEE and Odds Ratio Approach 354 20.3 Variance Estimation: One-Step EF–Bootstrap 356 20.4 Goodness-of-Fit Tests 357 20.4.1 Construction of Groups 358 20.4.2 Quasi-Score Test 358 20.4.3 Adjusted Hosmer–Lemeshow Test 360 20.5 Illustration Using NPHS Data 361 20.5.1 Parameter Estimates and Standard Errors 362 20.5.2 Goodness-of-Fit Tests 363 20.6 Summary 364 References 365 xii CONTENTS 21 A Latent Class Approach for Estimating Gross Flows in the Presence of Correlated Classification Errors 367 Francesca Bassi and Ugo Trivellato 21.1 Introduction 367 21.2 Correlated Classification Errors and Latent Class Modelling 368 21.3 The Data and Preliminary Analysis 371 21.4 A Model for Correlated Classification Errors in Retrospective Surveys 373 21.5 Concluding Remarks 378 Acknowledgements 379 References 379 22 A Comparison of Graphical Models and Structural Equation Models for the Analysis of Longitudinal Survey Data 381 Peter W. F. Smith, Ann Berrington and Patrick Sturgis 22.1 Introduction 381 22.2 Conceptual Framework 382 22.3 Graphical Chain Modelling Approach 383 22.4 Structural Equation Modelling Approach 384 22.5 Model Fitting 385 22.6 Results 386 22.7 Conclusions 390 Acknowledgements 391 References 391 Index 393[UseMoney=1]