A Propensity Score Approach towards Assessing Neig

Authors: Tusty ten Bensel Benjamin Gibbs Robert Lytle

Publish Date: 2014/08/30

Volume: 40, Issue: 2, Pages: 377-398

PDF Link

Abstract

Previous research pertaining to parole has primarily concentrated on individual level characteristics as predictors of parole success holding offender residency as a constant Recently there has been a growing realization that we need to understand how neighborhood contexts impact successful reintegration It is difficult to evaluate the increased risk associated with neighborhood characteristics on parole revocation because residency after prison is not randomly assigned Consequently methodological adaptations must be made to account for selection into disadvantaged neighborhoods To reduce overt bias we used two propensity score strategies matching and inverse probability of treatment weighting to determine the impact of parolee residence on parole outcome Our analytic strategies indicated a weaker effect for neighborhood context when compared to traditional models In essence we found that both techniques were effective at decreasing the threat to internal validity allowing for greater confidence in reporting an average treatment effect The implications of our findings are discussed as well as the utility of these analyses in criminal justice research

Keywords: