Adjustment for Covariates
This chapter introduces methods of statistical adjustment. Statistical adjustment is used to reduce the effects of confounders; or, more precisely, to infer what association would have been observed had there been no confounding. The main analytic methods for control of confounding include stratification, statistical modeling, and subgroup analysis. The chapter begins by focusing on stratification and regression analysis as methods of analysis for unmatched samples. It then describes methods for analyzing matched data. Although matching to control for confounding is done before the data are collected, it is critically important to apply statistical methods that account for the matching, because the analysis must correspond to the design to attain valid results. Subgroup analysis is briefly considered at the end of the chapter.
Oxford Scholarship Online requires a subscription or purchase to access the full text of books within the service. Public users can however freely search the site and view the abstracts and keywords for each book and chapter.
If you think you should have access to this title, please contact your librarian.