Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Analysis of proportion as the dependant variable

From   Ronan Conroy <>
To   "statalist" <>
Subject   Re: st: Analysis of proportion as the dependant variable
Date   Mon, 23 Jun 2003 10:25:10 +0100

on 22/06/2003 12:57 pm, Drora Fraser at wrote:

> We have data on proportion of children stunted at age 1 from 13
> locations, for each of 3 years for boys and girls.
The numbers of children in each group will vary, and it would be helpful to
take this into account. You can use -expand- to recreate a dataset with one
observation per child, by creating a variable that records the number of
children in each of the subcategories defined by sex, year, location and

Be careful however. If there are NO children in a category, -expand- will
not create zero observations, it will simply not expand that observation.

> We wish to examine the association between stnting and hospitalization,
> at the level of the location, contrlling for year, sex and other
> factors.

If you use Stata's -svy- commands, you can set location as a stratum
variable, allowing the prevalence of stunting to differ between location.
This should give you a clear run at a simple logistic model. I would try
putting in dummies for year2 and year3.

> Which of the 3 is the correct analysis: 1. multivariable regression, 2.
> XT-GEE regression with (group variable=location, link=log,
> family=poisson) 3. Poisson regression.
> We would appreciate an answer or a link to anyone who might be able to
> help us.

I would be slow to analyse the proportions themselves. Each proportion is
probably based on a different sample size, and so has a different degree of
precision. Reflating the data to person-level data will take this into

Ronan M Conroy (
Lecturer in Biostatistics
Royal College of Surgeons
Dublin 2, Ireland
+353 1 402 2431 (fax 2764)


This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom
they are addressed.
If you have received this email in error please notify the
originator of the message. This footer also confirms that this
email message has been scanned for the presence of computer viruses.

Any views expressed in this message are those of the individual
sender, except where the sender specifies and with authority,
states them to be the views of The Royal College Of Surgeons in Ireland.


*   For searches and help try:

© Copyright 1996–2019 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index