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

st: RE: subpops within subpops using svy commands

From   "Sayer, Bryan" <>
To   "'karen matsuoka '" <>, "' '" <>
Subject   st: RE: subpops within subpops using svy commands
Date   Mon, 9 Jun 2003 18:23:03 -0400

An unfortunate 'feature' in Stata is that you have to create a single
variable for each sub-pop you wish to analyze.  So you need to create a
variable something like

generate byte poordiabetics = pov < 1 & diabetes = 1

for example, and do this for every subpop you wish to analyze.  Be careful
of missing values.

Bryan Sayer
Statistician, SSS Inc. 

-----Original Message-----
From: karen matsuoka
Cc: Sayer, Bryan
Sent: 6/9/03 5:26 PM
Subject: subpops within subpops using svy commands

I would like to conduct a logistic regression using the svylogit command
on a sub-sample of a subsample of a complex survey sample. 

I understand that the appropriate way to work with sub-samples with
svy-commands is to use the subpop command. However, Stata seems to allow
only one subpop. 

What if I want to look at a subpopulation of a subpopulation (for
example, only individuals with incomes below federal poverty level who
also have diabetes)? If I were to use the subpop command to select those
people with diabetes, and then the "if" command to select only those
people with the income level I am interested in, will that work or will
that throw all of the standard errors off? 

Any advice would be greatly appreciated!


Get 25MB of email storage with Lycos Mail Plus!
Sign up today -- 
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index