Hi, see this link. I hope it helps. Best wishes, david

http://stackoverflow.com/questions/7732459/how-to-do-a-tukey-hsd-test-with-the-anova-command-car-package

These links might be useful for you:

https://www.youtube.com/watch?v=6-4mWkOgDtg

https://www.youtube.com/watch?v=9iL8v6A-gi0

If so please let me know.

best all luck

Gayatri

I wouldn't trust either TukeyHSD or HSD.test with a repeated measures design. I don't know, but I wouldn't trust them.

The approach I would recommend is to abandon the repeated measures framework in the car package.  And to use a mixed effects model with either the nlme or lme4 packages.

For post-hoc testing, you can then use the lsmeans package.

* * *

Attached is an example file with your data, and a couple of references.

I included the correlation option in the lme function, though I suspect with a data structure this simple, it does nothing. Normally you would use a ~ Time | Subject formula, but in this case the model didn't like that.

Using multcomp package after a mixed model fit by nlme is another option for post hoc comparisons.

Salvatore, the following works.

lme(Measurement ~ Time, random = ~1|Vaccine/Time,correlation = corAR1(form = ~ Time , value=-0.00936), data=Mice)

Dear

I agree with Salvatore