Principal components analysis

In statistics, principal components analysis was developed by Pearson in 1901 and may be used for analysis of a cohort study. The principal components analysis can only correct for confounding by independent variables that have been measured.

Software
Software available for meta-analysis includes:
 * R programming language:
 * HSAUR2 interactive package with a chapter containing sample demonstrations, "Principal Component Analysis: The Olympic Heptathlon" in "A Handbook of Statistical Analyses Using R".