BibTeX
@MISC{Department98improvingupon,
author = {Andrew Gelman Department},
title = {Improving Upon Probability Weighting for Household Size},
year = {1998}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
By comparing data from national telephone polls to Census figures on household size (number of adults in household), we find large differences between population and sample, even after weighting respondents proportional to household size. This presumably occurs because larger households are easier to reach and more likely to respond to the survey. If a user wishes to weight on household size, we recommend poststratification on known population proportions of household size instead. Keywords: opinion polls, poststratification, survey sampling, telephone surveys Introduction In survey sampling, inverse-probability weights are used to correct for unequal selection probabilities, and post-stratification weights are used to correct for known or expected discrepancies between the sample and the population (see, e.g., Kish, 1992). In this research note, we consider the effects of these adjustments for household size in telephone polling. In a survey in which households are sampled at random...
Keyphrases
household size upon probability weighting survey sampling national telephone poll research note telephone polling large difference population proportion unequal selection probability census figure opinion poll telephone survey introduction inverse-probability weight post-stratification weight
