Definition
This variable indicates the subsample number of the household randomly numbered from 0 to 99. Each subsample is nationally representative and preserves all stratification of the sample from which it is drawn.
This variable is a useful tool for carrying out the "subsample replicate" method of standard error estimation. This method involves dividing an IPUMS sample into 100 random subsamples and generating 100 subsample estimates for a given statistic. With these 100 "subsample replicate" estimates, researchers can measure a statistic's variation across all of the subsamples. Due to Census sample designs this method yields a more precise estimate of the standard error of a sample statistic than would be achieved through the application of a theoretical standard error formula.