Digitizing hand-written data with automated methods: A pilot project using the 1990 U.S. Census

Abstract

While the U.S. Census Bureau has microdata files from the 1960 through 1990 Decennial Censuses, respondent names were never digitized. Names from these censuses are only available in handwritten form on microfilm images of the original census manuscripts. In this paper, we document the 1990 Census Name Recovery Pilot (NRP) project, which was used to identify the most accurate and cost-effective means to recover respondent names, focusing on the example of the 1990 Census. In addition to describing the four stages of the project, the results of the NRP are presented in detail. The NRP showed that respondent names can be accurately digitized to support the integration of the 1960 through 1990 Censuses into the Census Bureau’s infrastructure of linked data from censuses, surveys, and administrative records.