R help - Jul 2007 - Multilevel package: Obtaining significance for waba within-group correlation?

Hello everyone,

I am employing the waba method from the multilevel package for obtaining 
a within-group correlation (Description: 
http://bg9.imslab.co.jp/Rhelp/R-2.4.0/src/library/multilevel/man/waba.html). 
Does anybody know a way or a calculation for obtaining a significance 
value for that correlation?

And another question: Does anybody know whether it is possible to save 
individual group-mean-centered values into a new variable?

I am quite new to R so please do apologize if either of these questions 
has an obvious answer that I didn't get.

Example:

 > package(multilevel)
 > data(bh1996)

 > waba(bh1996$HRS,bh1996$WBEING,bh1996$GRP)
[I would like to obtain the significance of the CorrW value]

Thanks for your help,
Bertolt

-- 
Bertolt Meyer
PhD Student
Institute of Psychology & Institute of Information Systems
Humboldt University Berlin

bmeyer at psychologie.hu-berlin.de

phone: +49-30-20939347
mobile: +49-179-4991046

web1: http://ioe-skillmap.hu-berlin.de
web2: http://amor.rz.hu-berlin.de/~h04440am

On 7/23/07, Bertolt Meyer <bmeyer@psychologie.hu-berlin.de>
wrote:
>
> And another question: Does anybody know whether it is possible to save
> individual group-mean-centered values into a new variable?

If data is in 'x' and grouping factor is 'group', use "x -
tapply(x, group,
mean)[group]"

For example:

group <- gl(2,10)
x <- rnorm(20)
x - tapply(x, group, mean)[group]




-- 
Christophe Pallier (http://www.pallier.org)

	[[alternative HTML version deleted]]