Message-ID: <Pine.LNX.4.21.0003071438300.1369-100000@boromir.ci.tuwien.ac.at>
Date: 2000-03-07T13:45:54Z
From: Torsten Hothorn
Subject: manova
In-Reply-To: <E12SK1v-0005OH-00@mailhub1.shef.ac.uk>
>
> >
> > After running a MANOVA test, I used univariate Fs in
> order to see
> > which IV(s) was contributing the difference.
> > However, I have been critised about it,
> > and the reason I was given is:
> >
> > "MANOVA allows you to explore multiple
> associations but
> > does not excuse you from tracing out the causal
> relationships
> > for each variable. This does not mean falling back on
> bivariate
> > correlation."
Yes, this is right! I had this problem too, and I used the following
article:
\bibitem{4}Kropf, Siegfried (1988). \textsl{Application of Multiple Test
Procedures to the Combination of Multivariate and Univariate Tests with
Varying Variable Sets}, Biometric Journal 30, pp. 461-470
If there are more then 4 variables, you'll run into great computational
troubles, I have some very problem specific code here to solve this for up
to 8 variables, if you are interested, drop me a mail.
Torsten
> >
> > Do you have any suggestion for me to find out
> mentioned causal
> > relationships?
> >
> > g. baris
> > dept of economics
> > university of sheffield
> >
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
> Send "info", "help", or "[un]subscribe"
> (in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
>
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._