Association for Public Policy Analysis & Management

This presentation will describe the methods and results of a systematic evaluation of the costs of the HOPE demonstration field experiment. HOPE was implemented in four jurisdictions and included a random controlled experiment to determine the impact of swift, certain, and fair responses to probation violations to reduce crime compared to a Probation As Usual (PAU) control group. For both the HOPE and PAU groups, we first identified the relevant domains of costs (e.g., violation hearings). For each domain, we then obtained data on the number of events (e.g., number of hearings) and cost per event (e.g., cost of a hearing) per probationer. A variety of data sources were used, including fidelity measures of a sample of records on the time that program staff performed certain activities, semi-structured interviews, the literature, and administrative records. We then combined the number of events with the cost per event to compute the average taxpayer cost per probationer across all domains and for each domain across the four HOPE sites. We also compared costs for HOPE and PAU to determine whether HOPE was cost effective.