Analyzing Complex Survey Data

by Eun Sul Lee

• ISBN13: 9780761930389
• ISBN10: 0761930388
• eBook ISBN(s): 9781506320434
• Edition: 2nd
• Format: Paperback
• Copyright: 2005-09-22
• Publisher: Sage Publications, Inc
• More Book Details

Summary

This book examines ways to analyze complex surveys, and focuses on the problems of weights and design effects. This new edition incorporates recent practice of analyzing complex survey data, introduces the new analytic approach for categorical data analysis (logistic regression), reviews new software and provides an introduction to the model-based analysis that can be useful analyzing well-designed, relatively small-scale social surveys.

Table of Contents

Series Editor's Introduction	v
Acknowledgments	vii
1. Introduction	1	(2)
2. Sample Design and Survey Data	3	(8)
Types of Sampling	4	(3)
The Nature of Survey Data	7	(2)
A Different View of Survey Data*	9	(2)
3. Complexity of Analyzing Survey Data	11	(11)
Adjusting for Differential Representation: The Weight	11	(3)
Developing the Weight by Poststratification	14	(3)
Adjusting the Weight in a Follow-Up Survey	17	(1)
Assessing the Loss or Gain in Precision: The Design Effect	18	(2)
The Use of Sample Weights for Survey Data Analysis*	20	(2)
4. Strategies for Variance Estimation	22	(17)
Replicated Sampling: A General Approach	23	(3)
Balanced Repeated Replication	26	(3)
Jackknife Repeated Replication	29	(6)
The Bootstrap Method	35	(1)
The Taylor Series Method (Linearization)	36	(3)
5. Preparing for Survey Data Analysis	39	(10)
Data Requirements for Survey Analysis	39	(2)
Importance of Preliminary Analysis	41	(2)
Choices of Method for Variance Estimation	43	(1)
Available Computing Resources	44	(3)
Creating Replicate Weights	47	(2)
Searching for Appropriate Models for Survey Data Analysis*	49	(1)
6. Conducting Survey Data Analysis	49	(29)
A Strategy for Conducting Preliminary Analysis	50	(2)
Conducting Descriptive Analysis	52	(5)
Conducting Linear Regression Analysis	57	(4)
Conducting Contingency Table Analysis	61	(4)
Conducting Logistic Regression Analysis	65	(4)
Other Logistic Regression Models	69	(5)
Design-Based and Model-Based Analyses*	74	(4)
7. Concluding Remarks	78	(2)
Notes	80	(3)
References	83	(5)
Index	88	(3)
About the Authors	91