Improved Method for Estimating Resident Nonimmigrant Population Size

• Activity Number: 660
• Type: Contributed
• Date/Time: Thursday, August 4, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Section on Government Statistics
• Author(s): Bryan C. Baker*+
• Companies: Department of Homeland Security
• Keywords: missing records ; administrative records ; administrative data ; immigration ; nonimmigrant ; population size

Abstract

The U.S. Dept. of Homeland Security matches administrative records of arrivals and departures together to estimate the size of the resident nonimmigrant population in the U.S. during a given period of time. The size is estimated, rather than measured, because the U.S. does not have strict departure control, and because not all visits have ended, which leaves the collection of departure records incomplete. The method used in the most recent DHS report synthesized departure dates by assuming that all visits within a given category were of average length. It synthesized departure dates for all arrivals, regardless of the presence or absence of a departure record, because longer, recent visits were disproportionately likely to be continuing visits (hence no departure record) and synthesizing departure dates only for arrivals with no matching departure would tend to replace longer visits with average length visits. This paper presents an alternative method that only synthesizes departure dates in the absence of a departure record, avoiding the bias associated with continuing visits by categorizing visits according to recentness and calculating estimates for each category separately.