Statalist


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

RE: st: pseudo R2s for Generalized Linear Models


From   "Maarten Buis" <M.Buis@fsw.vu.nl>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: pseudo R2s for Generalized Linear Models
Date   Fri, 20 Jul 2007 14:21:46 +0200

--- Ronan Conroy wrote:
> Another point to consider is that the catchphrase "percentage 
> of variance explained" assumes that the percentage of variance 
> explainable is 100%.

Yet another point is whether we would want to live in world 
where we can explain a 100% of the variance. When teaching I 
propose a model explaining whether or not someone is faithful 
to their partner. Typically some students than rebel against 
the idea that such behaviour can be explained and claim that 
it is their choice and their choice alone. However, when I 
show the results and the pseudo r square is pretty low, they 
correctly claim that that proves their point, but for the 
wrong reason. They say models with low R square "are bad", 
ergo the model I estimated is bad and should not have been 
estimated at all. The correct reasoning is that there is a 
lot of room for free choice (or other variables, or better 
measurement of the existing variables).  


-----------------------------------------
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/
-----------------------------------------



*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index