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

st: RE: Interpreting Poisson output

From   "Kieran McCaul" <>
To   <>
Subject   st: RE: Interpreting Poisson output
Date   Wed, 24 Jun 2009 10:53:16 +0800


Yes, but remember the result would be college males relative to no
college females.  Also, you are assuming that there is no multiplicative
interaction between college and sex.  In other words, the sex effect is
the same regardless of whether or not you went to college (and the
college effect is the same regardless of sex).

You should probably check that by fitting i.gender*

Kieran McCaul MPH PhD
WA Centre for Health & Ageing (M573)
University of Western Australia
Level 6, Ainslie House
48 Murray St
Perth 6000
Phone: (08) 9224-2701
Fax: (08) 9224 8009
If you live to be one hundred, you've got it made.
Very few people die past that age - George Burns

-----Original Message-----
[] On Behalf Of Data
Analytics Corp.
Sent: Wednesday, 24 June 2009 9:18 AM
To: Stata Listserve
Subject: st: Interpreting Poisson output

Hi Stata,

I have a simple question about interpreting Poisson output.  Suppose I 
have a simple model with counts of newspapers read as the dependent 
variable and gender and education as the independent variables.  The two

independent variables are both dummy coded so gender is just 1 = male/0 
= female and education is just 1 = college/0 = no college.  If I use

        poisson count gender education, irr

I get reasonable answers.  The gender coefficient exponentiated tells me

the change in the count for males relative to females.  Similarly for 
education.  What if I want to know males who went to college?  Do I just

exponentiate the sum of the two coefficients?  I think so.  Are there 
any things to watch for?




Walter R. Paczkowski, Ph.D.
Data Analytics Corp.
44 Hamilton Lane
Plainsboro, NJ 08536
(V) 609-936-8999
(F) 609-936-3733

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2022 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index