StATS: Profile analysis and MANOVA (April 18, 2005)

Someone asked me about profile analysis as alternative analysis to MANOVA (Multivariate Analysis of Variance). Typically you would use profile analysis when the outcome variables are measuring (more or less) the same thing, but possibly at different times or in different ways. You start by examining a profile of these measures, a graph that looks very similar to an interaction plot. You first test for parallelism by looking at a set of contrasts. If you accept the null hypothesis here, then you look to see if the profiles are flat, again using a contrast. Finally, if you accept that null hypothesis, you test whether the profiles are coincident (lie one on top of the other).

Both MANOVA and profile analysis have been replaced by better and more flexible approaches using a mixed model analysis of variance and/or a random effects regression model. I want to write a web page about mixed models and random effects models, but have not had the time to do this.

Further reading:

  1. www.ats.ucla.edu/stat/stata/faq/profile.htm
  2. socsci.colorado.edu/LAB/STATS/SPSS/spss1095.html

This page was written by Steve Simon while working at Children's Mercy Hospital. Although I do not hold the copyright for this material, I am reproducing it here as a service, as it is no longer available on the Children's Mercy Hospital website. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at Category: Mixed linear model.