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

Re: st: Resampling question

From   BRUNDAGE Thomas W <[email protected]>
To   [email protected], [email protected]
Subject   Re: st: Resampling question
Date   23 Apr 2003 14:12:37 -0700

Julia -

You could post-stratify (weight) the data to account for the over-response by
women, and then include the weight in the logit command.  This would be less
problematic than using a random process to toss "extra" records out of your


thomas brundage
Department of Human Services
Oregon Public Health
800 NE Oregon St., #827
Portland OR 97232

voice: 503-731-3448
FAX:   503-872-5398

>>> [email protected] 4/23/2003 1:17:00 PM >>>
The goal is to run a logit model with a dataset that is representative of
the distribution of males and females in a particular subject area.

The problem is that I am working with a sample with far more female than
male survey respondents. Of 215 randomly selected survey respondents, 86 are
male, 129 female. According to the female proportion in the population, I
would only include 33 female responses.

Is there a way (e.g. through bootstrapping or Monte Carlo simulation) that I
can keep the 86 male respondents in the sample and match them with a series
of 33 randomly selected female respondents from the total 129 female
respondents in the sample? Is it possible to run a logit model in this

Thank you in advance for any help.
Julia Koschinsky 

Julia Koschinsky
Project Director
PIN-NY -- Policy Information Network of New York State
Center for Women in Government & Civil Society
University at Albany
Draper Hall 303A
135 Western Avenue
Albany, NY 12222

Phone: 1 518-442-5781
Fax: 1 518-442-3877

Email: [email protected] 

*   For searches and help try:
*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index