[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Differences in regression slopes

From   Maarten buis <>
Subject   RE: st: Differences in regression slopes
Date   Wed, 20 Feb 2008 18:55:39 +0000 (GMT)

--- Barth Riley <> 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,

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

Sent from Yahoo! Mail - a smarter inbox

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index