--- Barth Riley <[email protected]> wrote:
> The standard approach is to perform separate analyses for each group
> and compare group-specific betas. In my case, I wanted to do this but
> with the interaction of two or more factors (e.g., gender by race). I
> can certainly see now that this is a more complex problem than the
> DIF literature would suggest!
At some point you need to prioritize which problems you are going to
solve. Doing these interactions is progres over just doing seperate
analyses and eyeballing the resulting parameters.
To add to the confusion: the method you propose is perfectly legitemate
as a so-called population average model (i.e. describing differences
between groups) but is not a causal model (group membership causes the
differences in probability). I nice describtion of the difference can
be found in chapter 13 of (Fitzmaurice et al. 2004) (the population
average model is called marginal model and causal model is called mixed
effects model)
Hope this helps,
Maarten
Fitzmaurice, Garrett M., Nan M. Laird, and James H. Ware (2004)
"Applied Longitudinal Analysis". Hoboken: Wiley.
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
__________________________________________________________
Sent from Yahoo! Mail - a smarter inbox http://uk.mail.yahoo.com
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/