Migration Status Gradients in Immigrant Poverty: A Comparison of Imputation Methods

Abstract

Research on the stratifying effects of migration status has increased sharply in the last two decades, although efforts have been hampered by the near absence of representative data that include detailed migration status measures. Researchers have developed various statistical and logical imputation methods that have produced widely varying estimates. In this article, we introduce a new indicator of migration status constructed from two federal surveys matched to the Social Security Administration's Numident file, a database that includes all citizens and legal residents of the United States. In models predicting poverty, our measure produces estimates comparable to those based on respondents’ own self-reports, in one federal survey, of their migration status. Both the administrative and survey-based measures produce poverty gradients that diverge from those produced by logic-based measures. Our findings contribute to mounting evidence of bias in the use of certain kinds of logic-based algorithms to impute migration status and demonstrate the promise of administrative record linkages in migration status research.