Sampling Weights and Regression Analysis

Sociological Methods and Research - Tập 23 Số 2 - Trang 230-257 - 1994
Christopher Winship1, Larry Radbill2
1Harvard University
2Joint Economic Committee of Congress

Tóm tắt

Most major population surveys used by social scientists are based on complex sampling designs where sampling units have different probabilities of being selected. Although sampling weights must generally be used to derive unbiased estimates of univariate population characteristics, the decision about their use in regression analysis is more complicated. Where sampling weights are solely a function of independent variables included in the model, unweighted OLS estimates are preferred because they are unbiased, consistent, and have smaller standard errors than weighted OLS estimates. Where sampling weights are a function of the dependent variable (and thus of the error term), we recommend first attempting to respecify the model so that they are solely a function of the independent variables. If this can be accomplished, then unweighted OLS is again preferred. If the model cannot be respecified, then estimation of the model using sampling weights may be appropriate. In this case, however, the formula used by most computer programs for calculating standard errors will be incorrect. We recommend using the White heteroskedastic consistent estimator for the standard errors.

Từ khóa


Tài liệu tham khảo

Amemiya, Takeshi, Advanced Econometrics

10.2307/2095230

Cochran, William G., Sampling Techniques

10.1080/01621459.1983.10478006

Efron, Bradley, An Introduction to the Bootstrap

Goldberger, Arthur S., A Course in Econometrics

Greene, William H., Econometric Analysis

Hanushek, Eric A., Statistical Methods for Social Scientists

10.2307/1912682

10.2307/2982065

Kish, Leslie, Survey Sampling

Kott, Phillip S., 1991, American Statistician, 45, 107, 10.1080/00031305.1991.10475779

Leamer, Edward E., Specification Searches: Ad Hoc Inference with Nonexperimental Data

Maddala, G. S., Econometrics

10.2307/1914121

10.2307/1912934

10.2307/1912526

White, Halbert, Asymptotic Theory for Econometricians

10.1146/annurev.so.18.080192.001551