The Marshall Project is a nonprofit newsroom covering the U.S. criminal justice system. Sign up for our newsletters to receive all of our stories and analysis. Any discussion of criminal justice policy inevitably includes the word “recidivism.” Usually more than once. Recidivism is the reoccurrence of crime among people known to have committed crimes before. At all levels of justice, from local probation offices to the U.S. Bureau of Prisons, if we judge the impact of interventions at all, we do so in part by measuring recidivism. In a report we published today with the Harvard Kennedy School, we conclude that recidivism is often the wrong measure. And using it exclusively to assess the quality of justice is like using a school’s dropout rate to measure the success of teachers—it may be pertinent, but it is inadequate and often misleading. First, recidivism is not reducible to the behavior of individuals. For example, an illegal act cannot become the basis for recidivism until it is reported to law enforcement. And many crimes, including half of all violent crimes, are never reported to police, according to the federal Bureau of Justice Statistics. Social context also matters. Naturally, law violators have different statistical probabilities of coming into contact with police depending on where they live. In America, where one lives is largely determined by income and demographics. Pound for pound, poor people and people of color are more likely to be surveilled by law enforcement and more likely to be among those considered recidivists. Rather than asking “what’s the recidivism rate?” we should ask an entirely different set of questions about justice interventions. Are we really helping people convicted of crimes to form better relationships with their families and their law-abiding friends? Are we helping them to advance their educational goals? Are they more likely to develop the skills and abilities required for stable employment? Are we helping them to respect others and to participate positively in the civic and cultural life of their communities? These questions are critical because they look beyond the “yes or no” of recidivism and focus on factors that we know moderate criminal behavior—social bonds, education attainment, employment; they all facilitate what researchers call “desistance,” or the process by which people learn to become law-abiding. To be sure, the difference is more than rhetorical. Focusing on desistance instead of recidivism leads justice systems to reorient their operations and their measurement of success. A desistance framework encourages justice agencies to promote and monitor positive outcomes. For example, when a person previously convicted of a violent crime is later convicted of theft, it represents an incidence of recidivism. However, it could also indicate desistance, a step away from crime. The same could be said of the burglar who’s arrested five times in one year but only arrested once the following year. Despite promising research on the potential for desistance-focused approaches to improve outcomes, community corrections agencies continue to rely on recidivism as the primary measure of their effectiveness. But if justice systems are to be truly correctional, policymakers should begin to hold them to more rigorous standards, including asking systems to measure what they promise to produce and not merely what they try to avoid. This would aid policymakers in ensuring that they are getting the most from limited resources. If we required our justice system to support the process of success instead of tracking single instances of failure, we just might be more effective at reducing crime and improving public safety. Jeffrey Butts is director of the Research and Evaluation Center at John Jay College of Criminal Justice. Vincent Schiraldi is a senior research scientist at the Columbia School of Social Work and co-founder of the Columbia Justice Lab.