5.5. Analysis in the Presence of Covariates

When data contain covariates that might affect repeated measures tests, it is important to design the analysis to account for the covariate effects. Consider these examples. First, suppose the initial measurements on the subjects before the treatments are applied provide the information relevant to the subsequent measurements. Second, suppose various blood pressure drugs are administered to groups of subjects, and blood pressure measurements are taken weekly. Also measured is the amount of sodium consumed by each individual the previous day. See Milliken (1989). Since the amount of sodium in the diet is known to affect blood pressure measurements, a better analysis may be possible by incorporating ...

Get APPLIED MULTIVARIATE STATISTICS: WITH SAS® SOFTWARE now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.