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

RE. st: how to explain Rho in heckprob

From   "John Reynolds" <>
To   <>, <>
Subject   RE. st: how to explain Rho in heckprob
Date   Thu, 11 Mar 2004 15:42:27 -0500

Jun Xu,
The change in sign & significance of rho may be due to better specification
of your model(s) OR to collinearity.  See references below, especially
Stolzenberg.  Also, scholars in sociology don't agree 100% on how to
interpret rho.  Some say it is meaningful and can be interpreted as the
*effect* of the tendency to be in the sample on the outcome, but others
disagree and insist that rho is simply accounting for a spurious association
between error terms due to unmeasured factors.  Good luck.
John Reynolds

Useful references in sociology:

Berk, Richard A. 1983. "An Introduction to Sample Selection Bias in
Sociological Data." American Sociological Review 48: 386-98.

Berk, Richard A., and Subhash C. Ray. 1982. "Selection Biases in
Sociological Data." Social Science Research 11:352-98.

Breen, Richard. 1996. Regression Models: Censored, Sample Selected, or
Truncated Data. Sage University Paper series on Quantitative Applications in
the Social Sciences, 07-111. Beverly Hills, CA: Sage.

Mirowsky, John, and John Reynolds. 2000. "Age, Depression, and Attrition in
the National Survey of Families and Households." Sociological Methods and
Research 28(4):476-504.

Stolzenberg, Ross M., and Daniel A. Relles. 1990. "Theory Testing in a World
of Constrained Research Design: The Significance of Heckman's Censored
Sampling Bias Correction for Nonexperimental Research." Sociological Methods
and Research 18: 395-415.

Stolzenberg, Ross M., and Daniel A. Relles. 1997. "Tools for Intuition about
Sample Selection Bias and Its Correction." American Sociological Review
62(3): 494-507.

Winship, Christopher, and Robert D. Mare. 1992. "Models for Sample Selection
Bias." Annual Review of Sociology 18: 327-50.

>I am using Stata's heckprob to estimate a heckit probit model. However, it 
>seems there is little available out there re explaining Rho, for example, 
>what's the meaning of Rho being positive or negative. I estimate four
>heckprob models, and as I add more variables in, the sign of Rho changes, 
>and its significance level varies too. Any input?

John Reynolds, PhD
Department of Sociology
Florida State University
Tallahassee, FL 32306-2270
850-644-2304 (fax)

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index