• Presentation

Response probability weight adjustments using logistic regression

Citation

Iannacchione, V. G., Milne, J. G., & Folsom, R. E. (1991, August). Response probability weight adjustments using logistic regression. Presented at 151st Annual Meeting of the American Statistical Association, Section on Survey Research Methods, Atlanta, GA.

Abstract

Most adjustments for survey nonresponse are made by adjusting the sampling weights of respondents in a way that accounts for the nonrespondents. For example, weightingclass adjustments are based on the assumption that sample members can be partitioned into cells, or weighting classes, within which the responses of nonrespondents, had they been obtained, would be similar to those of respondents. Within each weighting class, the inverse of the weighted response rate is applied to the sampling weights of respondents so that the adjusted weight sums of respondents reproduce the unadjusted weight sums over respondents and nonrespondents.