effect size for mixed models

Can anyone refer me to reference to determine the effect size for a mixed
model longitudinal design?

Thanks.

[Daniel Robertson]

Steve Raudenbush and colleagues developed a nice little Windows program
for power analysis of longitudinal and multi-level designs. Go to
<http://sitemaker.umich.edu/group-based/optimal_design_software> for the
software and documentation.

Charla Lopez wrote:
> Can anyone refer me to reference to determine the effect size for a mixed
> model longitudinal design?
>
> Thanks.
>
>

--
Daniel Robertson
Sr. Research and Planning Associate
Institutional Research and Planning
Cornell University
440 Day Hall, Ithaca NY 14853-2801
ph:607.255.9642 / irp.cornell.edu

On Mon, 2 Apr 2007 10:59:34 -0400, Daniel Robertson <[hidden email]>
wrote:

>Steve Raudenbush and colleagues developed a nice little Windows program
>for power analysis of longitudinal and multi-level designs. Go to
><http://sitemaker.umich.edu/group-based/optimal_design_software> for the
>software and documentation.
>
>Charla Lopez wrote:
>> Can anyone refer me to reference to determine the effect size for a mixed
>> model longitudinal design?
>>
>> Thanks.
>>
>>
>
>--
>Daniel Robertson
>Sr. Research and Planning Associate
>Institutional Research and Planning
>Cornell University
>440 Day Hall, Ithaca NY 14853-2801
>ph:607.255.9642 / irp.cornell.edu

Thanks - I am really looking to test the effect size of a mixed model that
I have already run, rather than to run a power analysis, which is what
Raudenbush's Optimal Design program seems to be tailored to.  Does anyone
know if it is kosher to use the Rosenthaland Rosnow (1991)conversion of the
t-statistic to cohen's d for the main effect?


formula for effect size of 2 groups =d = t(n1 + n2) / [sqrt(df)sqrt(n1n2)]


Thanks again.