[R] welch anova and post-hoc

[peter dalgaard]

On Jun 12, 2011, at 12:50 , eldor ado wrote:

> dear r community...
> 
> it loks like i won't be able to reach homogenity of variance for my
> dataset, so i end up with welch anova instead of regular anova.
> documentation on this test is rather scarce, so maybe someone here can
> enlighten me a bit:
> 
> - do i understand that no two-way implementation of the welch anova
> has been developed yet?

Has the THEORY been developed? For complete data, I suppose you can formulate it as a multivariate contrast test and employ Huynh-Feldt correction, but it is not quite the obvious generalization. Otherwise, formulate as a mixed model with error variance depending on treatment, but the degree-of-freedom adjustments are not implemented in lme.

> - is there a post-hoc test for welch anovas implemented in R?

Again, what does it mean? 

pairwise.t.test(..., pool.sd = FALSE, var.eq = FALSE) 

will do various pairwise comparison and allow simple corrections of the p value (Bonferroni and close relatives). 

> 
> thanks a lot,
> lukas kohl
> 
> ______________________________________________
> R-help at r-project.org 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.

-- 
Peter Dalgaard
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: pd.mes at cbs.dk  Priv: PDalgd at gmail.com