Statalist


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

RE: AW: st: Stability of gllamm results against linear transformation of independent variables


From   "Verkuilen, Jay" <JVerkuilen@gc.cuny.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: AW: st: Stability of gllamm results against linear transformation of independent variables
Date   Thu, 20 Dec 2007 11:19:53 -0500

Maarten Buis wrote:

[snip]

>>There are other operationalizations of democracy than those provided by
Freedom House. As a robustness check I would also look at:

http://www.cidcm.umd.edu/polity/data/
http://www.prio.no/page/Project_detail//9649/42472.html<<

(Warning: I have a nasty head cold today so you may be hearing the Nyquil talking....) 

As one of the authors of a widely cited review piece on democracy indicators (from my pre-psychometrics life), none of them are ideal. You are probably dealing with variation up near the ceiling since most of the cases will be listed as "democratic" already. 

This might well argue for you to use multiple indicators in the hope that there will be more variance if you set things up in GLLAMM not as a path model for your dependent variable but as an SEM, using, say, Polity, Vanhanen, ACLP, etc., as indicators of a latent variable "democratization". I could easily see making this latent variable dichotomous given the way that the indicators are likely to work, but you might be able to tell from looking at the raw data. 

GLLAMM is ideally suited to doing this sort of thing. 

Jay

<<winmail.dat>>




© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index