Chapter 11. Missing Data
Published Online: 8 MAR 2002
DOI: 10.1002/0470846283.ch11
Copyright © 2000 John Wiley & Sons, Ltd
Book Title
Quality of Life: Assessment, Analysis and Interpretation
Additional Information
How to Cite
Fayers, P. M. and Machin, D. (2002) Missing Data, in Quality of Life: Assessment, Analysis and Interpretation, John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/0470846283.ch11
Publication History
- Published Online: 8 MAR 2002
- Published Print: 18 APR 2000
ISBN Information
Print ISBN: 9780471968610
Online ISBN: 9780470846285
- Summary
- Chapter
Keywords:
- missing data;
- bias;
- mean imputation;
- hierarchical scales;
- regression imputation;
- Markov chain imputation;
- hot deck imputation;
- cold deck imputation;
- EM algorithm
Summary
This chapter describes problems that arise through missing QoL assessment data. Situations are outlined where values are missing from otherwise completed questionnaires or where entire forms are missing. The main difficulty with either type of missing data is the bias they may introduce at the analysis stage. QoL data that are “missing” through attrition because the patient has died, from that which could be anticipated but was not returned, are distinguished. How missing values may be estimated, often termed as imputed, to ease the statistical analysis are described, but stress that imputing values is no substitute for collecting “real” data.