Survey Methodology and Missing Data: Tools and Techniques for Practitioners

1.Introduction
2.Concept of survey and key survey terms2.1What is survey?2.2Five populations in surveys2.3Purpose of populations2.4Cross-sectional survey micro data2.5X variables, auxiliary variables in more details2.6Summary of the terms and the symbols of Chapter 22.7Transformations3.Designing a questionnaire and survey modes3.1What is questionnaire design?3.2One or more modes in one survey?3.3Questionnaire and questioning3.4Designing questions for the questionnaire3.5Developing questions for the survey3.6Satisficing3.7Straightlining3.8Examples on questions and scales4.Sampling principles and missingness mechanisms4.1Basic concepts, both for probability and non-probability sampling4.2Missingness mechanisms4.3Non-probability sampling cases4.4Probability sampling framework4.5Sampling and inclusion probabilities4.6Illustration of the stratified three-stage sampling4.7Basic weights of stratified three-stage sampling4.8Two types of sampling weights5.Design effects at sampling phase5.1DEFF due to clustering = DEFFc5.2DEFF due to varying inclusion probabilities = DEFFp5.3The entire design effect - DEFF, and the gross sample size5.4How to decide the sample size and allocate the gross sample into strata? 6. Sampling design data file 6.1 Principles of the sampling design data file 6.2 Test data used in several examples in the book 7.Missingness, its reasons and treatment7.1Reasons for unit nonresponse7.2Coding of item nonresponse7.3Missingness indicator and missingness rate7.4Response propensity models8.Weighting adjustments due to unit missingness8.1Actions of weighting and reweighting8.2Introduction to re-weighting methods8.3Post-stratification8.4Response propensity weighting8.5. Comparisons of weights in other surveys8.6Linear calibration8.7Non-linear calibration8.8Summary of all the weights9.Special cases in weighting9.1Sampling of individuals, estimates for clusters such as households9.2If analysis weights only are available but the proper weights are required9.3Sampling and weights for households, estimates for individuals or other lower level9.4Panel of two years10.Statistical editing10.1Edit Rules and ordinary checks10.2Other edit checks10.3Satisficing in editing10.4Selective editing10.5Graphical editing10.6Tabular editing10.7Handling screening data in editing10.8Editing not always completely done for public use data11.Introduction to statistical imputation11.1Imputation and its purpose11.2Targets for imputation should be specified clearly11.3What can be imputed due to missingness?11.4'Aggregate imputation'11.5Most common tools for missing item handling without proper imputation11.6Several imputations for the same micro data12.Imputation methods for single variables12.1Imputation process12.2.Imputation model12.3.Imputation task12.4.Nearness metrics of real-donor methods12.5.Post-Editing after the model-donor method possibly12.6Single and multiple imputation12.7Examples of Deterministic imputation methods for a continuous variable12.8Example of deterministic imputation methods for a binary variable12.9Example of the continuous variable when the imputation model is poor12.10Interval estimates13.Summary and key tasks of survey data cleaning14.Basic survey data analysis14.1'Survey instruments' in the analysis14.2Simple and demanding examples14.2.1The sampling weights vary much14.2.2Feeling about household's income nowadays with two types of weights14.2.3Examples based on the test data (Chapter 6)14.2.4Using sampling weights for cross-country survey data without country results14.2.5The PISA literacy scores14.2.6Multivariate linear regression with survey instruments14.2.7The binary regression model with logit link14.3Concluding remarks about the results based on simple and complex methodology