Evaluation of the Honest Opportunity Probation With Enforcement Demonstration Field Experiment (HOPE DFE)

The proposed research is an evaluation of the Honest Opportunity Probation with Enforcement (HOPE) program. This examination will occur through a rigorous Random Controlled Trial (RCT) in up to four (4) demonstration sites. At each site, 400 probationers will be randomly assigned to either the HOPE condition or Probation as Usual (PAU) for a total sample size of 1,600 probationers. This evaluation will focus on (1) an evaluation of HOPE implementation and program fidelity at the individual sites, (2) determining the effects of HOPE on outcomes through a rigorous outcome evaluation, and (3) an assessment of HOPE cost-effectiveness through a detailed economic analysis. Each of these will be described in detail below.

(1) A detailed cross-site HOPE protocol will be developed to assess implementation fidelity and lessons learned for replication and sustainability. Measures will be collected through stakeholder interviews, observations of court warning and violation hearings, and a review of court, probation, and HOPE project records. The PAU practices will also be assessed to provide a measure of any changes in practices among the control group to identify potential contamination.

(2) The outcome study will incorporate administrative data on appointment compliance, drug tests results, re-arrests, violations, revocations, and jail and prison days; interview data will be collected at study enrollment and six (6) months post-enrollment on measures that will facilitate understanding of the nature of individual change associated with HOPE. Quantitative analysis will be done in stages. First descriptive statistics followed by mixed effects analyses of variance to estimate random effects at the site level and fixed effects for HOPE compared to PAU. If preliminary analyses suggest site effects, then multiple-group methods will be used to estimate HOPE program effects within each site. If these analyses indicate probation officer effects then multilevel models will be used to analyze clustered data.

The evaluators will also conduct random drug tests on a subsample of HOPE and PAU probationers using oral swabs to provide a common measure of current drug use. HOPE probationers will also respond to periodic questionnaires during their daily call-in. These mini interviews will assess changes in attitudes over time in an intensive longitudinal design (ILD) using a growth curve and nonlinear dynamical systems.

(3) Process and outcome data will be combined with cost data to assess cost-effectiveness. The cost-effectiveness assessment will identify start-up costs, differential cost and savings impact per participating agencies, and the cost-effectiveness of HOPE for achieving desired non-monetized outcomes.
ca/ncf