Immigration and Unemployment in the United States: A Multilevel Analysis

Jing Li, PRO-ED, Inc.
Isao Takei, Texas A&M University

We examine the impact of the immigrant population on the work force of the United States. It is assumed that both immigration status and exposure to job markets with high proportions of immigrants operate to determine an individual's unemployment. We distinguish how immigration affects the likelihood of unemployment individually and contextually, and how it interacts with other factors. The data are derived from the 1995 Current Population Survey (CPS) and the State and Metropolitan Area Data Book 1997-98 (SMADB). First we use a logistic regression model to ascertain the effects of individual determinants. We then estimate three Hierarchical Generalized Linear Models (HGLM) to analyze direct and indirect effects of the individual-level and contextual-level predictors. The results show that a large percentage of the immigrant population greatly increases the whole population's likelihood of employment. However, a major part of this direct effect contributes to the immigrant's unemployment.