Introduction » Survey Design » Sampling Weights

With the introduction of the new cohorts in HRS 1998 the nature of the HRS sample changed dramatically, making it necessary to develop a revised set of sampling weights. Specifically the population of inference changed from American adults in the 1931 – 1941 or 1892 – 1923 birth cohorts to adults born prior to 1948. Furthermore we decided to revise weights for all past panel waves so as to use a consistent methodology. The new methodology involves post-stratifying each wave’s weights to the March CPS on the basis of the birth cohorts of the respondent and spouse and on the basis of respondent gender and race/ethnicity. These factors capture all the major characteristics of the sample design and provide consistent adjustments for sample attrition and mortality.

More...

Additional Information


Copyright © 2014 The Regents of the University of Michigan