Variance Estimation in U.S. Census Data from 1960-2010

Kathryn Coursolle, University of California, Los Angeles
Lara Cleveland, University of Minnesota
Steven Ruggles, University of Minnesota

Modern census microdata feature complex sample designs that clustered within households and incorporate stratification. Yet, researchers often calculate standard errors utilizing methods designed for simple random samples. Variance estimates can differ dramatically adjusting for complex survey design clustering and stratification relative to estimates assuming simple random sampling. Examining potential differences in variance estimation in recent IPUMS-USA samples is essential because US census microdata are among the most heavily used data sources for social, historical, demographic, and policy research. This project uses decennial census data from 1960-2000 and American Community Survey data from 2000-2010 to compare standard errors under the assumptions of simple random sampling to estimates which adjust for clustering and stratification, and subsample replicate weights for recent ACS data. We conclude by discussing potential implications of these techniques on statistical inference.