Applied Survey Data Analysis

"Anyone analyzing survey data, even once, should have a copy of this book. The book has something for everyone. It is a solid, yet accessible introduction to analyzing data from complex sample surveys (i.e., those with stratification and clustering), a statistical text of the highest caliber, and a reference for experienced analysts and statisticians. The authors are masterful instructors on the topic, and leaders in the field of survey methodology at the University of Michigan's world-renowned Institute for Social Research and Survey Research Center. Their profound understanding of the topic, and talent for describing it shines through vividly in the text. One of my favorite parts remains section 1.2 "A Brief History of Applied Survey Data Analysis", which is split into "Key Theoretical Developments" and "Key Software Developments". The historical context provided in those sections helps motivate the technical material that follows. My other favorite parts of this book are the presentations of analysis code and output from various programs, and their "Theory Boxes", which tie specific analysis steps and code to the statistical theory behind them. Among the numerous updates to this edition, I think readers will find the new content on model diagnostics and testing goodness-of-fit (GOF) to be extremely helpful, as this is an area of complex sample survey analysis that can be difficult to translate from standard regression analysis. Throughout, the authors make it a point to describe analyses in discrete steps that can help direct even the most complex analyses."
—Matt Jans, Senior Associate/Scientist, Abt Associates

"This is an excellent book to use for a graduate level applied statistics course teaching public health students how to analyze complex survey data. Each chapter is clearly written with a nice balance of theoretical background and practical guidance on survey data analytical issues as illustrated by many relevant real-data examples. All the chapters are nicely structured and coherently connected. The second edition includes a new chapter on the current approaches for analyzing longitudinal survey data, which will be instrumental for many practitioners trying to analyze this type of data. Several other new additions such as bootstrap methods of variance estimate, log-linear models, and fractional imputation methods for item missing data, are all important tools for an applied survey data analyst to learn and thus further enhance the scope and usability of this book. I would highly recommend this book to any instructor of an applied survey data analysis course and anyone who needs to analyze complex survey data to conduct their research."
—Luohua Jiang, Assistant Professor, Department of Epidemiology, University of California, Irvine

"The first edition of Applied Survey Data Analysis was my go-to reference and textbook for all types of analysis of complex survey data. This second edition is even better. The book maintains its practical analytic approach, understanding that all users of survey data want to know about methods that can be easily and appropriately implemented with major statistical analysis software, and that more sophisticated users of survey data want insights into why the methods work. In this edition, Heeringa, West, and Berglund have added an entire chapter about longitudinal data analyses (including multilevel models) using complex survey data, expanded the chapter on multiple imputation (including a discussion of fractional imputation), provided detailed examples for common complex survey data analyses that did not have them in the previous edition (such as the Pearson product-moment correlation coefficient and structural equation models), significantly and usefully expanded the discussion of model interpretation and evaluation, updated and expanded the examples to more recent data sets including the European Social Survey, and address the important perpetual question of the use of weights in regression models. All of the examples reflect the latest version of Stata (14+), including user-generated commands and other software programs where Stata falls short. The exercises in this edition also have been updated. This edition is hands-down the best book out there for an applied graduate-level course on complex survey data analysis or any user who wants to know how to obtain the correct design-based analyses."
— Kristen Olson, University of Nebraska-Lincoln

"I have been using Applied Survey Data Analysis in my sampling methods class for years. This is an ideal text for the course – it covers design, analysis and a host of extra topics that make it a one-stop shop for a survey methods curriculum. The new version really adds to the material without skimping or detracting from the integrity of the original. What I like best is that the authors try to cover the widest array of methods so they include survey data analysis using STATA, SAS, SUDAAN and WesVar PC all in one text. There are good books for designing samples and good books for analyzing public use datasets, but this book is the best of both worlds. Unlike most statistical texts, Heeringa and his co-authors write in a style that is very easy to read and avoids copious formulas and symbolic notation. This makes it a great option for someone who might want to delve into survey design and analysis where a full semester course is not available and a self-learning text would be desirable.

I think that the best measure of the quality of a text in a classroom setting is the prevalence of used copies I observe from semester to semester. While I see an abundance of used textbooks in the other classes I teach throughout the year I may only see one or two used copies of this one each term. This tells me that the students find the book just as helpful and useful a resource as I do."
—Paul Weiss, Senior Associate at the Rollins School of Public Health, Emory University

"This second edition is well written, organized, and also adds a new chapter on longitudinal complex sample survey data. This book is full of real data examples from several different analysis programs. It will be an excellent reference with 13 structured chapters giving an overview, basic structures and ideas, and real applications. Faculty, researchers, and graduate students working in survey data research and consulting could reasonably adopt this book as an introduction to survey knowledge and practical analysis skills from beginner to expert."
—Yoonsung Jung, Research Scientist/Statistician, Prairie View A&M University

Praise for the first edition:

"...the authors do an admirable job of striking a balance between statistical theory and practical advice and analysis. The authors provide excellent coverage of each aspect of the survey analysis process … This book is an excellent general resource and if the reader is left wanting on a topic the authors never fail to provide an ample set of citations and references to a wide variety of notable texts on the topic in question. … an excellent and helpful addition to the desk of any analyst, researcher, or student with a general background in statistics who is dealing with the special challenges and demands of complex survey data."
—Gregory Holyk, Journal of Official Statistics, Vol. 27, 2011

"Overall, the book is clearly written and easy to follow, and well equipped with real data examples and a book website. The program codes used in the example are also available, mostly written in Stata. I like the presentations with real survey examples and, in particular, the unified four-step approach to the regression analysis in different models. Anyone working on survey data analysis would find the book very helpful and instructive. The book website seems to be a good complement, with additional resources on this book."
—Jae-Kwang Kim, The American Statistician, November 2011

"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

"This book covers applied survey analysis, from survey design and weighting, to a complete list of techniques for analyzing survey data. As discussed in the preface, the motivation for the 2nd edition is to provide effective methods for teaching a complex subject as survey analysis. The book is appropriate for a graduate student with at least one applied statistics course and one statistical theory course, as well as any research scientist or survey practitioner desiring a complete and reliable survey reference. It is an excellent intermediate book, providing the opportunity to learn about complex survey design, weighting, and sampling error calculation models, as well as more advanced techniques, such as Poisson, negative binomial, multinomial logistic, survival analysis, longitudinal, and structural equation modelling, to name but a few. Overall, the text is an easy read containing a great balance of theory and applied explanations. ...The book and the corresponding resources impart years worth of experience, from both an industry perspective, as well as from teaching statistics and survey analysis. The wealth of knowledge is priceless to a graduate student, who may not cover half of the material in such detail until actually in the field."
-Stephanie A. Besser, ISCB June 2018

"Anyone analyzing survey data, even once, should have a copy of this book. The book has something for everyone. It is a solid, yet accessible introduction to analyzing data from complex sample surveys (i.e., those with stratification and clustering), a statistical text of the highest caliber, and a reference for experienced analysts and statisticians. The authors are masterful instructors on the topic, and leaders in the field of survey methodology at the University of Michigan's world-renowned Institute for Social Research and Survey Research Center. Their profound understanding of the topic, and talent for describing it shines through vividly in the text. One of my favorite parts remains section 1.2 "A Brief History of Applied Survey Data Analysis", which is split into "Key Theoretical Developments" and "Key Software Developments". The historical context provided in those sections helps motivate the technical material that follows. My other favorite parts of this book are the presentations of analysis code and output from various programs, and their "Theory Boxes", which tie specific analysis steps and code to the statistical theory behind them. Among the numerous updates to this edition, I think readers will find the new content on model diagnostics and testing goodness-of-fit (GOF) to be extremely helpful, as this is an area of complex sample survey analysis that can be difficult to translate from standard regression analysis. Throughout, the authors make it a point to describe analyses in discrete steps that can help direct even the most complex analyses."
—Matt Jans, Senior Associate/Scientist, Abt Associates

"This is an excellent book to use for a graduate level applied statistics course teaching public health students how to analyze complex survey data. Each chapter is clearly written with a nice balance of theoretical background and practical guidance on survey data analytical issues as illustrated by many relevant real-data examples. All the chapters are nicely structured and coherently connected. The second edition includes a new chapter on the current approaches for analyzing longitudinal survey data, which will be instrumental for many practitioners trying to analyze this type of data. Several other new additions such as bootstrap methods of variance estimate, log-linear models, and fractional imputation methods for item missing data, are all important tools for an applied survey data analyst to learn and thus further enhance the scope and usability of this book. I would highly recommend this book to any instructor of an applied survey data analysis course and anyone who needs to analyze complex survey data to conduct their research."
—Luohua Jiang, Assistant Professor, Department of Epidemiology, University of California, Irvine

"The first edition of Applied Survey Data Analysis was my go-to reference and textbook for all types of analysis of complex survey data. This second edition is even better. The book maintains its practical analytic approach, understanding that all users of survey data want to know about methods that can be easily and appropriately implemented with major statistical analysis software, and that more sophisticated users of survey data want insights into why the methods work. In this edition, Heeringa, West, and Berglund have added an entire chapter about longitudinal data analyses (including multilevel models) using complex survey data, expanded the chapter on multiple imputation (including a discussion of fractional imputation), provided detailed examples for common complex survey data analyses that did not have them in the previous edition (such as the Pearson product-moment correlation coefficient and structural equation models), significantly and usefully expanded the discussion of model interpretation and evaluation, updated and expanded the examples to more recent data sets including the European Social Survey, and address the important perpetual question of the use of weights in regression models. All of the examples reflect the latest vers