Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
NCTM or The Math Forum.


ggorjan
Posts:
18
Registered:
4/25/06


Difference between groups for many variables
Posted:
May 11, 2006 3:24 AM


Hello!
I often get data where there is more than one variable per unit, say height, body weight, etc. The task of the experiment is usually to test for the differences between groups. I usually performed univariate analysis with (generalized) linear model for each variable and performed inference (estimated difference with corresponding intervals and pvalues) for each variables separately. In a seminar dissusion of similar type of analysis a problem of multiple testing was brought up and people mentioned that one should lower alpha via Bonferroni correction. Can anyone please comment on this?
Is there any difference in approach to such type of posthoc adjustment between frequentist and Bayesian view of stats?
Additionally I was wondering if there is any better way to perform the analysis. I was thinking about:  multivariate analysis with linear models, where I would allow for correlation between variables  "true" multivariate analysis via Hotelling's T  I have never performed such ...
Thanks!
 Lep pozdrav / With regards, Gregor Gorjanc
 University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.unilj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc <at> bfro.unilj.si
SI1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europe fax: +386 (0)1 72 17 888
 "One must learn by doing the thing; for though you think you know it, you have no certainty until you try." Sophocles ~ 450 B.C. 



