Re: st: importance of R-squared in ANOVA

[Richard Williams <Richard.A.Williams.5@ND.edu>]

At 01:52 PM 7/15/2006, Inna Khousnoullina wrote:
> Dear statlist,
>
> I hope to find here an answer on my question.
> I run anova over my data and have significance values for F-test. but my
> r2 and adj. r2 are low (e.g. 0.0123- 0.0602). could it be a problem for
> my results? (my depvar is a categorical variable and I know that this
> could be a problem).
>
> Thank you very much.
>
> Inna Khousnoullina
If the dependent variable is categorical, why are you using 
Anova?  Something like mlogit or ologit might be more appropriate.

In any event, the key concern is with having a correctly specified 
model, and a correctly specified model may have a low R^2.  You may 
want to ask yourself, however, why  is the R^2 low?  If it is because 
the model is poorly specified, or the measures are low in 
reliability, then you may have a problem that you may or may not be 
able to do something about.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc 
*
*   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/