Probation revocation: A proportional hazards model of the conditioning effects of social disadvantage

Most studies find that offender's age, gender, ethnicity, prior arrest record, severity of the current offense and level of supervision significantly influence time to probation failure. There is little evidence to show that treatment interventions significantly affect either the likelihood of failure or the time to failure. We propose that an offender's prior record and lower education level - indicators of social disadvantage - directly affect the mean time to a probation revocation. Further, we suggest that social disadvantage may condition the effects of other offender characteristics, incident offense characteristics, and treatment intervention on failure time. Using a proportional hazards model of probation revocation, we find that intervention increases the risk of failure, as well as partial support for our hypothesis of the conditioning effect of offender's social disadvantage.