[R] Can one use correlations (or R^2) as data for an ANOVA?

From: Jose Quesada <quesada_at_gmail.com>
Date: Thu 10 Aug 2006 - 02:25:36 EST

Dear all,

This may seem obvious for some of you, but it got me thinking. Can one use correlations (or R^2) as data for an ANOVA?

The case in hand: I have several models fitting the same data (individual fits per subject). The different models fitting the same guy will produce different R^2, I was wondering if these could be used as dependent variable in a repeated measures design where the repeated factors are the different models.

For some reason, this "analysis of variance on 'variance explained' as data rings a bell as something not methodologically correct. On the other hand, R^2 may be distributed in a way that fulfill of the ANOVA assumptions.

Is it an aberration?

Thanks a lot in advance,

R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Thu Aug 10 02:34:30 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Thu 10 Aug 2006 - 04:20:01 EST.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.