Adjusted* Hazard Ratios for Crowding and Other Predictors of Earlier Parole Violations.

Summary of stratified recurrent events Cox regression analysis for prison crowding and other covariates predicting parole violation rates.* Results for prison crowding (and for other predictors of recidivism) were adjusted for: sex, race, mental health status, prior offense type, parole region, age first incarcerated, age when released, number of prior imprisonments, number of prior serious offenses, sex offender status, level of supervision, three-striker status, parole officer sex, and parole officer race. A Cox stratified procedure was used to adjust for covariates not meeting the proportional hazards assumption as well as those that did meet this assumption. These variables are sex offender status, parole region, age first incarcerated, age when released, number of prior imprisonments, and number of prior serious offenses.Adjusted* Hazard Ratios for Crowding and Other Predictors of Earlier Parole Violations.