[R] post-hoc test (glht?) which takes treatment into account not just explanatory variable overall

[Bert Gunter]

David:

1. Sounds like you want to test the **interaction** of song type and treatment.

2. Does glht  work ( S3 vs S4 class issues) with lmer fits?

3. If it works, does glht give sensible results? (see Doug Bates's
discussions on this issue).

4. Probably best to direct all such questions to the R-sig-mixed-models list.

Cheers,

Bert

On Fri, Apr 22, 2011 at 10:01 AM, Ellis, David <de223 at exeter.ac.uk> wrote:
> Hi R helpers!
>
> I have used a glht as a post-hoc test on an lmer with:
> -2 treatments (A & B)
> -1 categorical explanatory variable (song type)
> -1 response variable (latency to respond)
>
> I wanted to make comparisons between the categorical variables depending on treatment.
>
> At the moment the glht simply returns comparisons of each of the (3) categorical explanatory variables with each other overall i.e. there is no distinction between the two treatments.
>
> Does anyone know of a post-hoc test that will give me comparisons of the 3 explanatory variables compared with each other for each treatment as well as comparing the treatments against each other (basically one that takes treatment into account and gives more results)
>
> Does this make sense?
>
> Thanks
> Dave
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> 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.
>



-- 
"Men by nature long to get on to the ultimate truths, and will often
be impatient with elementary studies or fight shy of them. If it were
possible to reach the ultimate truths without the elementary studies
usually prefixed to them, these would not be preparatory studies but
superfluous diversions."

-- Maimonides (1135-1204)

Bert Gunter
Genentech Nonclinical Biostatistics