3rd Edition

Design and Analysis

  • Available for pre-order. Item will ship after November 30, 2021
ISBN 9780367279509
November 30, 2021 Forthcoming by Chapman and Hall/CRC
680 Pages 60 B/W Illustrations

USD $79.95

Prices & shipping based on shipping country


Book Description

"The level is appropriate for an upper-level undergraduate or graduate-level statistics major. Sampling: Design and Analysis (SDA) will also benefit a non-statistics major with a desire to understand the concepts of sampling from a finite population. A student with patience to delve into the rigor of survey statistics will gain even more from the content that SDA offers. The updates to SDA have potential to enrich traditional survey sampling classes at both the undergraduate and graduate levels. The new discussions of low response rates, non-probability surveys, and internet as a data collection mode hold particular value, as these statistical issues have become increasingly important in survey practice in recent years… I would eagerly adopt the new edition of SDA as the required textbook." (Emily Berg, Iowa State University)

What is the unemployment rate? What is the total area of land planted with soybeans? How many persons have antibodies to the virus causing COVID-19? Sampling: Design and Analysis, Third Edition shows you how to design and analyze surveys to answer these and other questions. This authoritative text, used as a standard reference by numerous survey organizations, teaches the principles of sampling with examples from social sciences, public opinion research, public health, business, agriculture, and ecology. Readers should be familiar with concepts from an introductory statistics class including probability and linear regression; optional sections contain statistical theory for readers familiar with mathematical statistics.

The third edition, thoroughly revised to incorporate recent research and applications, includes a new chapter on nonprobability samples—when to use them and how to evaluate their quality. More than 200 new examples and exercises have been added to the already extensive sets in the second edition.

SDA’s companion website contains data sets, computer code, and links to two free downloadable supplementary books (also available in paperback) that provide step-by-step guides—with code, annotated output, and helpful tips—for working through the SDA examples. Instructors can use either R or SAS® software.

  • SAS® Software Companion for Sampling: Design and Analysis, Third Edition by Sharon L. Lohr (2022, CRC Press)
  • R Companion for Sampling: Design and Analysis, Third Edition by Yan Lu and Sharon L. Lohr (2022, CRC Press)

Table of Contents


1. Introduction
 Guidance from Samples                             
 Populations and Representative Samples                   
 Selection Bias                                  
 Convenience Samples                           
 Purposive or Judgment Samples                    
 Self-selected Samples                           
 What Good are Samples with Selection Bias?             
 Measurement Error                               
 Questionnaire Design                              
 Sampling and Nonsampling Errors                       
 Why Use Sampling?                               
 Advantages of Taking a Census                     
 Advantages of Taking a Sample Instead of a Census         
 Chapter Summary                                

2. Simple Probability Samples
 Types of Probability Samples                          
 Framework for Probability Sampling                      
 Simple Random Sampling                            
 Sampling Weights                                
 Confidence Intervals                               
 Using Statistical Software to Analyze Survey Data              
 Determining the Sample Size                          
 Systematic Sampling                              
 Randomization Theory for Simple Random Sampling*            
 Model-Based Theory for Simple Random Sampling*             
 When Should a Simple Random Sample Be Used?              
 Chapter Summary                                

3. Stratified Sampling
 What is Stratified Sampling?                          
 Theory of Stratified Sampling                         
 Sampling Weights in Stratified Random Sampling              
 Allocating Observations to Strata                       
 Proportional Allocation                         
 Optimal Allocation                            
 Allocation for Specified Precision within Strata            
 Which Allocation to Use?                        
 Determining the Total Sample Size                   
 Defining Strata                                  
 Model-Based Theory for Stratified Sampling*                 
 Chapter Summary                                

4. Ratio and Regression Estimation
 Ratio Estimation in Simple Random Sampling                
 Why Use Ratio Estimation?                       
 Bias and Mean Squared Error of Ratio Estimators          
 Ratio Estimation with Proportions                   
 Ratio Estimation Using Weight Adjustments             
 Advantages of Ratio Estimation                    
 Regression Estimation in Simple Random Sampling             
 Estimation in Domains                             
 Ratio Estimation with Stratified Sampling                  
 Model-Based Theory for Ratio and Regression Estimation*         
 A Model for Ratio Estimation                      
 A Model for Regression Estimation                   
 Differences Between Model-Based and Design-Based Estimators   
 Chapter Summary                                

5. Cluster Sampling with Equal Probabilities
 Notation for Cluster Sampling                         
 One-Stage Cluster Sampling                          
 Clusters of Equal Sizes: Estimation                   
 Clusters of Equal Sizes: Theory                     
 Clusters of Unequal Sizes                        
 Two-Stage Cluster Sampling                          
 Designing a Cluster Sample                           
 Choosing the psu Size                          
 Choosing Subsampling Sizes                       
 Choosing the Sample Size (Number of psus)              
 Systematic Sampling                              
 Model-Based Theory for Cluster Sampling*                  
 Estimation Using Models                        
 Design Using Models                           
 Chapter Summary                                
Contents ix

6. Sampling with Unequal Probabilities
 Sampling One Primary Sampling Unit                     
 One-Stage Sampling with Replacement                    
 Selecting Primary Sampling Units                   
 Theory of Estimation                          
 Designing the Selection Probabilities                  
 Weights in Unequal-Probability Sampling with Replacement     
 Two-Stage Sampling with Replacement                    
 Unequal-Probability Sampling Without Replacement            
 The Horvitz–Thompson Estimator for One-Stage Sampling     
 Selecting the psus                            
 The Horvitz–Thompson Estimator for Two-Stage Sampling     
 Weights in Unequal-Probability Samples                
 Examples of Unequal-Probability Samples                  
 Randomization Theory Results and Proofs*                 
 Model-Based Inference with Unequal-Probability Samples*         
 Chapter Summary                                

7. Complex Surveys
 Assembling Design Components                        
 Building Blocks for Surveys                       
 Ratio Estimation in Complex Surveys                 
 Simplicity in Survey Design                       
 Sampling Weights                                
 Constructing Sampling Weights                     
 Self-Weighting and Non-Self-Weighting Samples            
 Estimating Distribution Functions and Quantiles               
 Design Effects                                  
 The National Health and Nutrition Examination Survey          
 Graphing Data from a Complex Survey                    
 Univariate Plots                             
 Bivariate Plots                              
 Chapter Summary                                

8. Nonresponse
 Effects of Ignoring Nonresponse                        
 Designing Surveys to Reduce Nonresponse                  
 Two-Phase Sampling                              
 Response Propensities and Mechanisms for Nonresponse          
 Auxiliary Information for Treating Nonresponse            
 Methods to Adjust for Nonresponse                  
 Response Propensities                          
 Types of Missing Data                          
 Adjusting Weights for Nonresponse                      
 Weighting Class Adjustments                      
 Regression Models for Response Propensities             
 Poststratification Using Weights                    
 Raking Adjustments                           
 Steps for Constructing Final Survey Weights             
 Advantages and Disadvantages of Weighting Adjustments      
 Deductive Imputation                          
 Cell Mean Imputation                          
 Hot-Deck Imputation                          
 Regression Imputation and Chained Equations            
 Imputation from Another Data Source                 
 Multiple Imputation                           
 Advantages and Disadvantages of Imputation             
 Response Rates and Nonresponse Bias                     
 Calculating and Reporting Response Rates              
 What is an Acceptable Response Rate?                
 Nonresponse Bias Assessments                     
 Chapter Summary                                

9. Variance Estimation in Complex Surveys
 Linearization (Taylor Series) Methods                     
 Random Group Methods                            
 Replicating the Survey Design                     
 Dividing the Sample into Random Groups               
 Resampling and Replication Methods                     
 Balanced Repeated Replication (BRR)                 
 Creating and Using Replicate Weights                 
 Generalized Variance Functions                        
 Confidence Intervals                               
 Confidence Intervals for Smooth Functions of Population Totals   
 Confidence Intervals for Population Quantiles             
 Chapter Summary                                

10. Categorical Data Analysis in Complex Surveys
 Chi-Square Tests with Multinomial Sampling                 
 Testing Independence of Factors                    
 Testing Homogeneity of Proportions                  
 Testing Goodness of Fit                         
 Effects of Survey Design on Chi-Square Tests                 
 Contingency Tables for Data from Complex Surveys         
 Effects on Hypothesis Tests and Confidence Intervals         
 Corrections to Chi-Square Tests                        
 Wald Tests                                
 Rao–Scott Tests                             
 Model-Based Methods for Chi-Square Tests              
 Loglinear Models                                 
 Loglinear Models with Multinomial Sampling             
 Loglinear Models in a Complex Survey                 
 Chapter Summary                                

11. Regression in Complex Surveys
 Model-Based Regression in Simple Random Samples             
 Regression with Complex Survey Data                    
 Point Estimation                             
 Standard Errors                             
 Multiple Regression                           
 Regression Using Weights versus Weighted Least Squares      
 Using Regression to Compare Domain Means                 
 Interpreting Regression Coefficients from Survey Data            
 Purposes of Regression Analyses                    
 Model-Based and Design-Based Inference               
 Survey Weights and Regression                     
 Survey Design and Standard Errors                  
 Mixed Models for Cluster Samples                   
 Logistic Regression                               
 Calibration to Population Totals                        
 Chapter Summary                                

12. Two-Phase Sampling
 Theory for Two-Phase Sampling                        
 Two-Phase Sampling with Stratification                    
 Ratio and Regression Estimation in Two-Phase Samples           
 Two-Phase Sampling with Ratio Estimation              
 Generalized Regression Estimation in Two-Phase Sampling     
 Jackknife Variance Estimation for Two-Phase Sampling           
 Designing a Two-Phase Sample                         
 Two-Phase Sampling with Stratification                
 Optimal Allocation for Ratio Estimation                
 Chapter Summary                                

13. Estimating the Size of a Population
 Capture–Recapture Estimation                         
 Contingency Tables for Capture–Recapture Experiments       
 Confidence Intervals for N                        
 Using Capture–Recapture on Lists                   
 Multiple Recapture Estimation                         
 Chapter Summary                                

14. Rare Populations and Small Area Estimation
 Sampling Rare Populations                           
 Stratified Sampling with Disproportional Allocation         
 Two-Phase Sampling                           
 Unequal-Probability Sampling                     
 Multiple Frame Surveys                         
 Network or Multiplicity Sampling                    
 Snowball Sampling                            
 Sequential Sampling                           
 Small Area Estimation                             
 Direct Estimators                            
 Synthetic and Composite Estimators                  
 Model-based Estimators                         
 Chapter Summary                                

15. Nonprobability Samples
 Types of Nonprobability Samples                       
 Administrative Records                         
 Quota Samples                              
 Judgment Samples                            
 Convenience Samples                           
 Selection Bias and Mean Squared Error                    
 Random Variables Describing Participation in a Sample       
 Bias and Mean Squared Error of a Sample Mean           
 Reducing Bias of Estimates from Nonprobability Samples          
 Estimate the Values of the Missing Units               
 Measures of Uncertainty for Nonprobability Samples         
 Nonprobability vs Low-response Probability Samples            
 Chapter Summary                                

16. Survey Quality
 Coverage Error                                  
 Measuring Coverage and Coverage Bias                
 Coverage and Survey Mode                       
 Improving Coverage                           
 Nonresponse Error                                
 Measurement Error                               
 Measuring and Modeling Measurement Error             
 Reducing Measurement Error                      
 Sensitive Questions                            
 Randomized Response                          
 Processing Error                                 
 Total Survey Quality                              
 Chapter Summary                                

A Probability Concepts Used in Sampling
A Probability                                    
A Simple Random Sampling with Replacement             
A Simple Random Sampling without Replacement           
A Random Variables and Expected Value                    
A Conditional Probability                             
A Conditional Expectation                            
A Exercises                                     


View More



Sharon L. Lohr, the author of Measuring Crime: Behind the Statistics, has published widely about survey sampling and statistical methods for education, public policy, law, and crime. She is a Fellow of the American Statistical Association and an elected member of the International Statistical Institute, and has received the Gertrude M. Cox, Morris Hansen, and Deming Awards. Formerly Dean’s Distinguished Professor of Statistics at Arizona State University and a Vice President at Westat, she is now a statistical consultant and writer.


"In summary, the revisions of SDA hold value for practitioners, educators, and students. The level is appropriate for an upper-level undergraduate or graduate-level statistics major. SDA will also benefit a non-statistics major with a desire to understand the concepts of sampling from a finite population. A student with patience to delve into the rigor of survey statistics will gain even more from the content that SDA offers. The updates to SDA have potential to enrich traditional survey sampling classes at both the undergraduate and graduate levels. The new discussions of low response rates, non-probability surveys, and internet as a data collection mode hold particular value, as these statistical issues have become increasingly important in survey practice in recent years. I have personally used past editions of SDA as a resource in my research and work. I am therefore comfortable recommending that students purchase the SDA. I expect that many of them will find SDA to be a useful reference, even beyond their coursework…The revision of SDA is not tied to a specific software language. The updated online supplements allow one to easily use SDA in conjunction with either SAS or R. I would eagerly adopt the new edition of SDA as the required textbook." (Emily Berg, Iowa State University) 

"I believe that this book now shines above the competing texts. The examples and problems are updated and more relatable to today’s student. I believe that the practice analyzing real data will give students a competitive edge on today’s job market. Once this is published, I will absolutely adopt this textbook in my course. (Truly, I have been dying for an updated book for Sampling Theory!)" (Samantha Seals, University of West Florida)

"I love Lohr’s text on this subject. This book should be adopted as many of the new additions I have reviewed here, including Chapter 15 on nonprobability samples, really elevate it and allow it to retain its relevance amid many changes and advances in our field." (Trent D. Buskirk, Bowling Green State University)

"I think this is by far the best book on survey sampling at the undergraduate level. It is the perfect balance between theoretical and practical. It has an excellent set of exercises, and great suggested additional readings. I use it in my courses and recommend it to everyone…Excellent idea to expand discussion of nonprobability samples. That is very practically relevant." (Elaine Zanutto, University of Pennsylvania)