Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: Modeling % data

From   Marlis Gonzalez Fernandez <>
To   "" <>
Subject   RE: st: Modeling % data
Date   Wed, 22 Sep 2010 15:43:11 -0400

Thanks for the information Maarten.  I think that using glm as you described is the best way given my data.  Now, I am not familiar with this stata command... will I interpret the coefficiants the same was as a regular linear regression?


-----Original Message-----
From: [] On Behalf Of Maarten buis
Sent: Wednesday, September 22, 2010 11:52 AM
Subject: RE: st: Modeling % data

--- On Sep 22, 2010, at 8:03 AM, Marlis Gonzalez Fernandez wrote:
> My outcome variable is a % (% error in a language test).  We do have  
> many 0 and 100.  I need to be able to do a multiple regression to  
> adjust for known predictors of the variable vs. the predictors of  
> interest.
> It was suggested that I use qreg.  I've done so and it seems to  
> work. 

This all depends on how close your dependent variable gets to the boundaries of 0% and 100%. If the data stays well within the range of
20%-80% than I would have no problem using either -qreg- or just regular
-reg-. However, when you have observations that get close to these 
boundaries, you'll probably want to take them into account. For that 
there is a whole suite of commands available, which I discussed at the
last German Stata Users' Group meeting:

(You'd have to look at your variable as an proportion rather than a 
percentage. But that is trivial, just divide by 100.)

Hope this helps,

Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen


*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index