Using Administrative Records to Study Mobility: The Case of the Housing Assistance Supply Experiment

摘要

This paper describes a statistical procedure developed to analyze the mobility experience of households who participated in an experimental test of housing allowances. Although the procedure was designed for administrative records, it is generally applicable to other longitudinal microdata that are 'censored' by short periods of observation. The procedure's application here shows the major role played by eligibility requirements in the mobility of housing allowance recipients. It also highlights a potential inferential problem that is widespread in the use of administrative records. Program requirements may cause self-selection of participants according to (in this case) predisposition to move. Self-selection may therefore bias comparisons with general populations. (Author)