Statalist


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

st: Re:


From   Richard Williams <Richard.A.Williams.5@ND.edu>
To   statalist@hsphsun2.harvard.edu, statalist@hsphsun2.harvard.edu
Subject   st: Re:
Date   Tue, 26 Aug 2008 09:43:35 -0500

At 09:03 PM 8/25/2008, Sampaio, Marcelo wrote:
Dear all

I am a reserch fellow and I use stata to my analysis. I would like to know what we should do when in a multiple logistic analysis we have a output saying that the variable was dropped by estimability or dropped because it is constant. How can we fix it and how much it affects the analysis?

Thanks

Marcelo Sampaio
Resech Fellow
As Martin notes, if the variable really is a constant, it can't do anything for your model. However, if you don't believe it should be a constant, you might double-check your sample selection procedures or the cases that are being lost because of missing data. For example, a dummy variable for gender will wind up being a constant if your sample has been limited to women. This might occur (without you realizing it) if, say, one of the variables in your model was a question asked only of women. Particularly in a large survey data set with lots of skip patterns, it can be a little tricky sometimes to keep track of just who exactly was asked the questions.


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

*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* 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