Taking a practical approach that draws on the authors' extensive teaching, consulting, and research experiences, Applied Survey Data Analysis provides an intermediate-level statistical overview of the analysis of complex sample survey data. It emphasizes methods and worked examples using available software procedures while reinforcing the principles and theory that underlie those methods. After introducing a step-by-step process for approaching a survey analysis problem, the book presents the fundamental features of complex sample designs and shows how to integrate design characteristics into the statistical methods and software for survey estimation and inference. The authors then focus on the methods and models used in analyzing continuous, categorical, and count-dependent variables; event history; and missing data problems. Some of the techniques discussed include univariate descriptive and simple bivariate analyses, the linear regression model, generalized linear regression modeling methods, the Cox proportional hazards model, discrete time models, and the multiple imputation analysis method.
The final chapter covers new developments in survey applications of advanced statistical techniques, including model-based analysis approaches. Designed for readers working in a wide array of disciplines who use survey data in their work, this book also provides a useful framework for integrating more in-depth studies of the theory and methods of survey data analysis. A guide to the applied statistical analysis and interpretation of survey data, it contains many examples and practical exercises based on major real-world survey data sets. Although the authors use Stata for most examples in the text, they offer SAS, SPSS, SUDAAN, R, WesVar, IVEware, and Mplus software code for replicating the examples on the book's website: http://www.isr.umich.edu/src/smp/asda/
Table of Contents:
Applied Survey Data Analysis: Overview
Introduction
A Brief History of Applied Survey Data Analysis
Example Data Sets and Exercises
Getting to Know the Complex Sample Design
Introduction
Classification of Sample Designs
Target Populations and Survey Populations
Simple Random Sampling: A Simple Model for Design-Based Inference
Complex Sample Design Effects
Complex Samples: Clustering and Stratification
Weighting in Analysis of Survey Data
Multistage Area Probability Sample Designs
Special Types of Sampling Plans Encountered in Surveys
Foundations and Techniques for Design-Based Estimation and Inference
Introduction
Finite Populations and Superpopulation Models
Confidence Intervals for Population Parameters
Weighted Estimation of Population Parameters
Probability Distributions and Design-Based Inference
Variance Estimation
Hypothesis Testing in Survey Data Analysis
Total Survey Error and Its Impact on Survey Estimation and Inference
Preparation for Complex Sample Survey Data Analysis
Introduction
Analysis Weights: Review by the Data User
Understanding and Checking the Sampling Error Calculation Model
Addressing Item Missing Data in Analysis Variables
Preparing to Analyze Data for Sample Subpopulations
A Final Checklist for Data Users
Descriptive Analysis for Continuous Variables
Introduction
Special Considerations in Descriptive Analysis of Complex Sample Survey Data
Simple Statistics for Univariate Continuous Distributions
Bivariate Relationships between Two Continuous Variables
Descriptive Statistics for Subpopulations
Linear Functions of Descriptive Estimates and Differences of Means
Exercises
Categorical Data Analysis
Introduction
A Framework for Analysis of Categorical Survey Data
Univariate Analysis of Categorical Data
Bivariate Analysis of Categorical Data
Analysis of Multivariate Categorical Data
Exercises
Linear Regression Models
Introduction
The Linear Regression Model
Four Steps in Linear Regression Analysis
Some Practical Considerations and Tools
Application: Modeling Diastolic Blood Pressure with the NHANES Data
Exercises
Logistic Regression and Generalized Linear Models (GLMs) for Binary Survey Variables
Introduction
GLMs for Binary Survey Responses
Building the Logistic Regression Model: Stage 1, Model Specification
Building the Logistic Regression Model: Stage 2, Estimation of Model Parameters and Standard Errors
Building the Logistic Regression Model: Stage 3, Evaluation of the Fitted Model
Building the Logistic Regression Model: Stage 4, Interpretation and Inference
Analysis Application
Comparing the Logistic, Probit, and Complementary Log-Log GLMs for Binary Dependent Variables
Exercises
GLMs for Multinomial, Ordinal, and Count Variables
Introduction
Analyzing Survey Data Using Multinomial Logit
Regression Models
Logistic Regression Models for Ordinal Survey Data
Regression Models for Count Outcomes
Exercises
Survival Analysis of Event History Survey Data
Introduction
Basic Theory of Survival Analysis
(Nonparametric) Kaplan–Meier Estimation of the Survivor Function
Cox Proportional Hazards Model
Discrete Time Survival Models
Exercises
Multiple Imputation: Methods and Applications for Survey Analysts
Introduction
Important Missing Data Concepts
An Introduction to Imputation and the Multiple Imputation Method
Models for Multiply Imputing Missing Data
Creating the Imputations
Estimation and Inference for Multiply Imputed Data
Applications to Survey Data
Exercises
Advanced Topics in the Analysis of Survey Data
Introduction
Bayesian Analysis of Complex Sample Survey Data
Generalized Linear Mixed Models (GLMMs) in Survey Data Analysis
Fitting Structural Equation Models to Complex Sample Survey Data
Small Area Estimation and Complex Sample Survey Data
Nonparametric Methods for Complex Sample Survey Data
References
Appendix: Software Overview
About the Author :
Steve G. Heeringa is a research scientist in the Survey Methodology Program, the director of the Statistical and Research Design Group in the Survey Research Center, and the director of the Summer Institute in Survey Research Techniques at the University of Michigan's Institute for Social Research. Brady T. West is a doctoral student and research assistant in the Survey Research Center at the University of Michigan's Institute for Social Research. He is also a statistical consultant in the Center for Statistical Consultation and Research. Patricia A. Berglund is a senior research associate in the Youth and Social Indicators Program and Survey Methodology Program in the Survey Research Center at the University of Michigan's Institute for Social Research.
Review :
The book is well-written by authors who have over 60 years of combined teaching and consultation experience in survey methodology and research techniques. It is excellent for reference, with 12 structured chapters coherently organised, providing intermediate-level statistical overview of techniques used in analysing complex survey data. ! It provides analysts with a framework of how to plan and conduct analysis of survey data, familiarise with terminologies used and understand common complex sample design features of clustering, stratification and weighting. ! it is an excellent reference book for Stata users and the accompanying website provides useful resources and updated information. I feel that the book seamlessly links theory with practical applications of the statistical methods and helps the reader to develop an understanding of the framework of thinking required to effectively analyse complex survey data sets. ! --E.C. Abraham, AQMeNtion Newsletter, April 2011 ! there is a wealth of instruction here. The writing style is expansive, keeping mathematics in check, and the material is well organized clearly into appropriate sections. I think that the book would serve any budding survey practitioner well: armed with the knowledge and practical skills covered herein, plus some real-life experience of course, one could reasonably claim to be well qualified in the subject. --International Statistical Review (2010), 78, 3
