Multivariate data analysis for outcome studies.

The use of multivariate statistical techniques for analyzing the complex data often gathered in outcome studies is discussed. The multivariate analysis of variance (MANOVA) is suggested for multiple group studies common to outcome studies. This technique can be utilized for a large number of specific research designs whenever multiple outcome measures are collected. MANOVA offers two specific advantages over more familiar univariate approaches: it presents better control over Type 1 error rates while preserving statistical power, and it allows more thorough analysis of complex data.