NLME and grouping functions

["Roberts, J. Kyle" <jkrobert@bcm.tmc.edu>]

Dear All,

 

I want to correlated the predicted values from a "lme" run, and the original outcome values from the original dataset.  The caveat is that I want to do it honoring the grouping structure.

 

So far, I have done this:

new<-split(data.frame(my.data$outcome, predict(lme.out)), my.data$group)

new.2<-lapply(new, cor)

 

The problem is that I now have a list in which each part of the list has the following:

 

$"1":
          X1        X2
X1 1.0000000 0.9327732
X2 0.9327732 1.0000000

$"2":
          X1        X2
X1 1.0000000 0.9449112
X2 0.9449112 1.0000000
. . . etc.

 

So here's my question:  How do I just get the correlation between the two variables and not have a correlation matrix?  The reason that I ask is because I will need to then unlist these correlations to include in another function I have built.

 

Thanks for the help,

Kyle

 

***************************************

J. Kyle Roberts, Ph.D.

Baylor College of Medicine

Center for Educational Outreach

One Baylor Plaza, MS:  BCM411

Houston, TX   77030-3411

713-798-6672 - 713-798-8201 Fax

jkrobert@bcm.edu

***************************************