Using Weights to Estimate Population Parameters from Survey Records

Daniel A. Relles

Published 1981

Explains an inexpensive and convenient procedure for obtaining nearly unbiased estimates of a population parameter [y] from a biased sample with only a small loss in the efficiency of the estimates. The usual procedure prescribed by statistical theory is to regress survey response variable [y] on explanatory variables whose population means can be reliably estimated from sample selection records. Those means are then plugged into the regression prediction formula. The simpler procedure is based on weighted means of [y]. The weighted-means procedure is applied to HASE data with good results. Provided the weights (1) adequately reproduce the population means of important predictor variables for [y] and (2) yield estimates with acceptably small variances, the simpler procedure is recommended for wide use in estimating population parameters from survey research data.

Document Details

Citation

Chicago Manual of Style

Relles, Daniel A., Using Weights to Estimate Population Parameters from Survey Records. Santa Monica, CA: RAND Corporation, 1981. https://www.rand.org/pubs/notes/N1136.html.
BibTeX RIS

This publication is part of the RAND note series. The note was a product of RAND from 1979 to 1993 that reported miscellaneous outputs of sponsored research for general distribution.

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit www.rand.org/pubs/permissions.

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.