[R-sig-ME] mcmcpvalue and contrasts

Ken Beath kjbeath at kagi.com
Fri Feb 29 23:25:15 CET 2008

On 28/02/2008, at 4:42 AM, Hank Stevens wrote:


> I have found, however, that it is often, yea, even more often, the  
> case that one factor, A, has a large independent effect, and that  
> another factor, B, moderates the effect of A to a small, albeit  
> detectable,  degree. In these cases, it makes biological sense to  
> discuss the "independent" effect of A. Whether one cloaks this in an  
> "average" effect of A, or states that the effect of A is "at least  
> b0, and as high as b0+b1, given some value of B."  I understand the  
> caution you express above -- I just did't want to see the baby go  
> along with the bath water.

There is nothing wrong with calculating an average effect, just that  
it needs to be explicit that it is for a certain population. Most  
programs allow for calculating combinations of parameter estimates,  
but I can't see anything to do this directly in R, so maybe it's  
expected that the user works out the linear algebra. One thing that  
helps is to reparameterise the model, so rather than fitting an  
interaction, the estimates are for the effect of A for the levels of  
B, so that the calculations are for a weighted sum.


More information about the R-sig-mixed-models mailing list