R help - May 2011 - routine for dependent correlation test with stratified random sample

Dear R-List, 
 
I would like to have a large number of stratified random subsamples drawn from
my dataframe and automatically test for correlation differences in every
subsample.
 
Let this be my dataframe
 
df<-data.frame(group=c(rep(1,5),rep(2,5),rep(3,5)),a=c(3,4,5,6,3,4,5,4,5,4,1,2,1,2,1),b=c(1,2,3,4,5,3,4,3,4,5,6,5,6,2,3),c=c(2,2,3,3,5,1,1,6,6,5,6,1,1,2,1))
 
Then I would like to have n subsamples with one row out of each group, e.g.
 
> df
   group a b c
1      1 3 1 2
2      1 4 2 2
3      1 5 3 3
4      1 6 4 3
5      1 3 5 5
6      2 4 3 1
7      2 5 4 1
8      2 4 3 6
9      2 5 4 6
10     2 4 5 5
11     3 1 6 6
12     3 2 5 1
13     3 1 6 1
14     3 2 2 2
15     3 1 3 1
 
>df.sub1
group  a  b  c
1         3  1  2
2         5  4  1
3         1  3  1
 
>df.sub2
group  a  b  c
1         4  2  2
2         4  3  6
3         2  2  2
 
etc.
 
And then test, if the correlation ab is significantly higher than the
correlation ac.
 
I managed to perform the test "manually" using r.test from the
"psych" package, however I did not succeed in doing it automatically,
i.e. I had to do cor(df.sub) for all subsamples an put the values manually into
the r.test-code (which is very time consuming if you have to do it 100 times).
Is there a nice way to combine the stratified subsampling with a code that can
do the r.test with dataframe input directly (I mean without me entering all
correlations ab, ac, bc manually)?
 
Thank you for any hint!
Alain 
	[[alternative HTML version deleted]]