Criminology and Public Policy

B. Criminology and Criminal Justice

For some individuals there is a powerful uneasiness between those who identify themselves with the traditional (sociological) roots of criminology and its search for understanding crime and those who associate themselves with criminal justice and the search for the right policies. In particular, concern has grown around criminology’s perceived attempt to establish itself as a distinct discipline, severing its ties with sociology and other more established fields of inquiry. Joachim Savelsberg and Robert Sampson (2002), for example, expressed concern that as criminology has tried to assert its intellectual independence (and establish itself as a separate discipline), it has lost much of its academic credibility. According to Savelsberg and Sampson, part of this lost credibility can be attributed to the field’s reliance on government funding and the concomitant reliance on state definitions and ideas. Savelsberg and Sampson argued that the study of criminology is at its best when attached to another discipline and led by intellectual tradition and ideas instead of by government priorities and dollars. James Austin (2003) similarly suggested that criminology’s irrelevance can be attributed to a lack of knowledge. Although he did not explicitly advocate a return to sociological roots, Austin argued that criminologists have little “good science” to offer policymakers, in large part because of the decline of scientific methods, unbridled speculation, researcher bias, and the heavy hand of government funders in criminological research.

Other individuals see a perfect melding between criminology and criminal justice as fields of inquiry. The best illustration of this melding is the “what-works” movement and its academic sibling, evidence-based criminology. The what-works movement draws most of its momentum from the now-infamous “nothing works” proclamation made by Robert Martinson and his colleagues in a seminal article published in the mid-1970s (Miller, 1989). Martinson et al. subjected research findings in the area of correctional rehabilitation to meta-analysis, an analytical method used to summarize research findings and isolate the size of effects across accumulated research studies. They used a primitive meta-analytic technique and, on the basis of their findings, argued that there was little evidence that rehabilitative programming had any appreciable effect. Although Martinson et al. did not offer a vision for what might work in place of rehabilitation, the take-away message became that we had been too “soft” on crime, and the gap left by rehabilitation was quickly filled with a full complement of new get-tough approaches to the problem of crime. Partly in reaction to the influence of Martinson et al.’s “nothing works” article, more recent criminologists have sought to provide better evidence as to “what works,” offering prescriptions for each of the major subareas of criminal justice (crime prevention, policing, juvenile justice, and corrections, among others).

In the late 1990s, the U.S. Congress asked for a comprehensive evaluation of program effectiveness in preventing crime. Congress’s request led the National Institute of Justice to commission a study of what works in crime prevention. The study, conducted by Lawrence Sherman and colleagues (1999), involved a review of more than 500 program impact evaluations. The researchers concluded that, in terms of crime prevention, some programs appeared to be working, some clearly did not work, and others showed promise.There were a number of programs for which the jury remained out, because the impact evaluations reviewed were not sound enough to allow the researchers to draw valid conclusions. Sherman et al. ultimately recommended that the Department of Justice primarily fund program evaluations seeking to address what works, particularly in high-crime urban areas.

The what-works paradigm draws on an evidence-based criminological approach. Evidence-based criminology requires that high-quality evaluation research form the basis of policy or practice and is perhaps best exemplified by The Campbell Collaboration, a group of interdisciplinary social scientists who undertake systematic reviews of research regarding the effectiveness of various social policies and practices. The Campbell Collaboration’s systematic reviews are designed to help inform both policy and practice through pulling together and synthesizing research findings to advance our understanding of best practices.

Despite the growing popularity of evidence-based criminology and the what-works paradigm, debate remains over what should constitute “evidence” and how high the bar should be set before the social science of criminology and criminal justice might be in a position to inform public policy. Some scholars believe the bar should be set very high and that only research that relies on true experimental methods (with random assignment to experimental and control groups) meets the bar. Truly experimental research is understandably hard to come by in a field where random assignment is at best difficult to achieve and, in some cases, not possible. Consider, for example, policy-driven research examining the effectiveness of a diversion program for serious offenders. Randomly assigning some offenders to this noncustodial program and others to prison would raise ethical concerns and present some rather complicated logistical challenges.

Other scholars think that research that has been subjected to the scrutiny of others in the field through the peer review process has the capacity to inform public policy. This broader view recognizes that there are limits to what we currently know and that there are limitations to the research that criminologists conduct but advocates for a more fluid approach in which research constantly informs the process. As knowledge grows and changes, so might policy.

Browse criminal justice research papers or view criminal justice research topics.