| Abstract: | The Social Security Administration (SSA) has, from its beginnings, recorded the race and ethnicity provided by those who apply for a Social Security card. Although some of these data are eventually used in published tabulations when persons file for benefits, problems with the data prevent a larger selection of published tables. These problems stem from: incomplete internal SSA computer processing; changes in the racial coding schemes over time; and missing codes for younger cohorts of applicants. In spite of these problems, more data can be shared with the public. This article shows how matching administrative files and using statistical techniques make it possible to associate a race/ethnicity code with the great majority of persons receiving a payment under the Supplemental Security Income (SSI) program, a means-tested program for persons who are aged or disabled. The article follows a 1-percent sample of SSI recipients through several steps in an attempt to develop a race code. This approach can provide data for the next several years on the race of all SSI recipients, as well as data on race and ethnicity for recipients under age 40. Beyond the next few years, these techniques will become less useful, and other methods will be needed. SSA is in the process of revising its standards for classification of federal data on race and ethnicity. The census for year 2000 will include coding changes. Other federal agencies will be given as long as January 2003 to comply with the new guidelines. |
