Book contents
- Frontmatter
- Contents
- Preface
- 1 Sampling methods
- 2 Weighting
- 3 Statistical effects of sampling and weighting
- 4 Significance testing
- 5 Measuring relationships between variables
- Appendix A Review of general terminology
- Appendix B Further reading
- Appendix C Summary tables for several common distributions
- Appendix D Chapter 2 mathematical proofs
- Appendix E Chapter 3 mathematical proofs
- Appendix F Chapter 4 mathematical proofs
- Appendix G Chapter 5 mathematical proofs
- Appendix H Statistical tables
- References
- Index
3 - Statistical effects of sampling and weighting
Published online by Cambridge University Press: 18 August 2009
- Frontmatter
- Contents
- Preface
- 1 Sampling methods
- 2 Weighting
- 3 Statistical effects of sampling and weighting
- 4 Significance testing
- 5 Measuring relationships between variables
- Appendix A Review of general terminology
- Appendix B Further reading
- Appendix C Summary tables for several common distributions
- Appendix D Chapter 2 mathematical proofs
- Appendix E Chapter 3 mathematical proofs
- Appendix F Chapter 4 mathematical proofs
- Appendix G Chapter 5 mathematical proofs
- Appendix H Statistical tables
- References
- Index
Summary
This chapter presents a subject often overlooked in statistical books: sampling and weighting methods may have a very strong and usually adverse effect on the statistical tools used in the evaluation of sample-based data, such as the standard error and confidence limits. Simple random sampling is rarely used in practice, except in limited or artificial situations. Even if it is used, the final sample may still turn out to be biased and may require some weighting to remove obvious imbalances. As a result, the usual simple formulae for standard errors and confidence limits become inadequate in most surveys and may lead to wrong conclusions.
The estimation of the variance of an estimate lies at the heart of any attempt to assess the confidence limits of that estimate or to apply any significance test to it. The term ‘variance’ is frequently misused and has a very precise and specific meaning here. The term is sometimes used loosely where ‘variation’ or ‘variability’ would be appropriate. It is also used elsewhere to mean a difference between an observed value and a reference or expected value. But in the context of statistics it is a measure of how individual observations vary around their mean value.
- Type
- Chapter
- Information
- Statistics for Real-Life Sample SurveysNon-Simple-Random Samples and Weighted Data, pp. 79 - 139Publisher: Cambridge University PressPrint publication year: 2006