In recent years, meta-analysis has been widely adopted as a means of informing sound policy decisions. Meta-analyses have been used successfully in evaluating the effectiveness of medical interventions, particularly the results of new drug trials. The goal of this type of study is to draw a general conclusion from the results of many clinical trials on the same drug—which often produce very different results—to determine whether the drug is safe enough and effective enough to be made available to the public. Over the past two decades, meta-analysis has become increasingly common in the study of programs and policies designed to reduce crime.
Recently, researchers have begun to combine meta-analyses with cost-benefit analysis. Meta-analysis answers the question: does the intervention produce positive outcomes? Cost-benefit analysis moves beyond this question to ask whether an effective intervention creates enough of a benefit to make it worth the cost. Thus, the combined meta-analysis and cost-benefit analysis attempt to answer the question: How do we make streets safer while spending less money?
DCPI has developed an empirical model that combines meta-analysis and cost-benefit analysis to begin to answer this question. DCPI has created a single-stage cost-benefit model using Bayesian processes to test whether the expected outcomes of implementing a policy or combination of policies in Washington, D.C., is worth the investment. The DCPI model will incorporate both the costs of delivering services in Washington, D.C., and the benefits of those services to District citizens. In particular, the DCPI model will incorporate savings to District citizens from reductions in the risk of becoming a victim of crime.
DCPI researchers will work closely with criminal justice agencies, community groups, and the public to inform transparent, evidence-based decisions about anti-crime spending. In the future, the model can be used to make evidence-based decisions when policymakers are confronted with difficult choices between successful programs when the resources to fund those programs are limited.
The Costs and Benefits of Functional Family Therapy for Washington, D.C.
This cost-benefit analysis of implementing a Functional Family Therapy (FFT) program in the District of Columbia indicates that the benefits are likely to outweigh the costs. The analysis employed an innovative statistical method that enables policymakers to assess the range of possible costs and benefits associated with specific evidence-based programs designed to prevent crime and recidivism. Results indicate that there is a 66 percent chance that an FFT program serving 150 juveniles will yield benefits exceeding its costs.
The Costs and Benefits of Electronic Monitoring for Washington, D.C.
This policy brief summarizes the second DCPI cost-benefit analysis employing an innovative statistical method that enables policymakers to assess the range of possible costs and benefits associated with specific evidence-based programs designed to prevent crime and recidivism. This particular study forecasted the costs and benefits of implementing an Electronic Monitoring program in the District. The analysis found an 80 percent chance that an EM program serving 800 people will yield benefits exceeding its costs.
The Costs and Benefits of Community-Based Substance Abuse Treatment in the District of Columbia
This policy brief summarizes the first of many Urban Institute cost-benefit analyses employing an innovative statistical method that enables policymakers to assess the range of possible costs and benefits associated with specific evidence-based programs designed to prevent crime and recidivism. This particular study examined the costs and benefits of the District of Columbia’s Community-Based Substance Abuse Treatment (CBSAT) program. The analysis found a 55 percent chance, on average, that the CBSAT program serving 150 people will yield benefits exceeding its costs.
Adult Criminal Justice Case Processing in Washington, DC
This report describes adult criminal case processing and disposition in Washington, DC. In general, this analysis finds that the District follows national patterns with respect to charging practice and sentencing. Defendants in the District are slightly more likely to be charged with serious person crimes than are cases nationally, and slightly less likely to be charged with a property crime. Defendants in the District are slightly more likely than other large urban counties to receive probation, and slightly less likely to be sentenced to jail, prison, or long prison sentences, though sentences for serious person offenses are longer.
A Bayesian Meta-Analysis of Drug Court Cost-Effectiveness
This report is the final report of the first year of DCPI’s efforts to develop a meta-analytic tool to predict the costs and benefits of criminal justice programs implemented in the District. It presents a Bayesian meta-analysis of drug courts and pairs these with nonparametric benefit estimates derived from the Multi-site Adult Drug Court Evaluation and simulation-based costs based on the Superior Court Drug Intervention Program. It provides a detailed methodological discussion and compares a number of commonly used meta-analytic methods, weighing the advantages and disadvantages of each, before presenting a relatively uncommon multilevel model which generalizes these methods. Conclusions discuss methodological and policy implications.
Meta-Analytic Cost-Benefit Estimates
This presentation was delivered during the DCPI Research Advisory Board meeting on April 20, 2011. It discusses the meta-analytic and cost-benefit methods used to develop predictive cost-benefit estimates of programs potentially implemented in the District. It covers the goals of the project and the methods used before demonstrating the utility of the model with a comparative meta-analysis of six criminal justice programs for adult offenders. It also discusses future plans for the model.
A Bayesian, Meta Cost-Benefit Model
DCPI staff presented initial results from its Bayesian statistics-driven empirical model to Brookings Institution staff. This presentation outlined the motivations for developing the model and its analytic strategy. Staff used a meta-analysis of drug courts to demonstrate the model's capabilities; specifically, what it would cost to implement a drug court in the District of Columbia and the expected benefits of its implementation. Further, the model describes the optimal characteristics a Washington, D.C. drug court should possess in order to maximize benefits. These include the number of participants, the nature of their current and previous offenses, and their motivations.