Skip navigation.
Home | About Us | Contact Us
Rick Scott, Governor
Florida Department of Corrections, Secretary Michael D. Crews

Florida Department of Corrections
Michael D. Crews, Secretary

Introduction

Recidivism is the reoccurrence of criminal behavior by offenders after intervention by the criminal justice system. Studies of criminal behavior consistently show that some offenders return to crime after arrests, convictions, punishments, and correctional program participation. Those who work in and study the justice system analyze recidivism to understand why the system sometimes does not prevent or reduce subsequent crime.

Relevance

In fiscal year 2002 alone, the Florida Department of Corrections spent over $1.3 billion to imprison more than 73,000 inmates. Over 26,000 of these inmates were released during the year as their sentences ended. Nearly 44% of inmates admitted to prison that year had previously been in prison in Florida.

This major budget expense, the public concern about crime, and the return of many inmates to prison make recidivism an important public policy issue. To address the issue properly, researchers must use the best data available and the most appropriate statistical tools to:

  • measure recidivism
  • account for what influences it
  • identify what may reduce it
  • determine how much it may be lowered.

This report is designed to help readers understand how recidivism data is collected and analyzed so they can obtain and use Florida prison inmate recidivism statistics appropriately.

Key Concepts

Recidivism is typically measured and reported as a rate—an inmate recidivism rate, for example, as the percentage of inmates released during a specific period who commit a new crime within a certain time following their release. Three essential components for calculating rates determine what those rates mean, whether rates are comparable, and the quality of program evaluations that rely on recidivism rates:

  • the cohort - the group of inmates released from prison during a defined time period,
  • the failure - the measure(s) that identify whether and when recidivism occurred,
  • the follow-up - the time elapsed since release from prison.

For release cohort members with the same follow-up period, the recidivism rate is the percentage calculated by dividing the number who fail by the total offenders in the cohort:

recidivism rate at follow-up =   number of recidivists from cohort within follow-up
total number of releases in cohort

Recidivism results from different studies may not be comparable, because they use different cohorts, failure events, and follow-up periods. Their results may also not be comparable if the data sources they used differ in important ways that affect any of these components.

Existing Knowledge

Certain facts about prison inmate recidivism rates are confirmed by studies of large release cohorts.

  1. Broader measures of the failure event yield higher rates than narrower measures—so rearrest rates exceed reconviction rates which exceed reimprisonment rates. This is logically expected because not all arrests result in convictions, nor do all convictions result in incarceration.

  2. Inmates released from state prisons have relatively stable recidivism rates—especially for reconviction—over time and across jurisdictions. These rates can be expected to appear in any analysis of large samples of released inmates (see Technical Appendix).

  3. Certain factors influence recidivism rates, especially several basic inmate characteristics beyond or largely outside the control of corrections agencies. Such factors related to higher recidivism rates include:
    • being male, young, or non-white;
    • with low educational achievement, prior recidivism, or serious criminal history;
    • having prison misbehavior, high security custody needs, or shorter prison stays.

Cautions

That these factors correlate with recidivism rates does not mean they cause crime. Yet the prevalence of these correlations in existing research does mean that studies of recidivism should not ignore their statistical effects. An evaluation may claim that a program reduces recidivism, but that effect is only credible if the study accounts properly for these factors' influences.

One statistic readily available from most corrections agencies is similar to and sometimes confused with a recidivism rate. Often called a "reincarceration rate" in prison systems, this statistic is the percentage of the current prison population that was in prison before. This is not a recidivism rate because these returned inmates came back after different follow-up periods: some after six months, others after six years. Also, this reincarceration rate depends on the number of first-time inmates in the prison population—if the number of first-time inmates grows faster than the number of recidivists in the population, the "reincarceration rate" will decline, even though the number of recidivists is also growing. Such "rates" do indicate a percentage of inmates who have returned, but they are not comparable or appropriate to interchange with recidivism rates.

Purpose

Reducing recidivism is an important public policy objective. Both the public and policymakers ask a reasonable question: what can be done to keep known criminals from reoffending? From a public policy perspective, the key to dealing with the costs recidivists impose is to apply what is known about recidivism. Properly applying the findings of recidivism studies requires understanding:

Such information helps answer the policy, scientific, and practical questions: what, if anything, reduces recidivism, how much, for which kinds of offenders, and under what conditions?

This document establishes recidivism rates for inmates released from Florida Department of Corrections facilities and reports the effects on those rates of factors that are readily measurable from Department data sources at the time an inmate is released. The Department does not have data on other factors appearing after release that may also influence rates including: unemployment, substance abuse, housing stability, etc. Nevertheless, this report's findings should guide evaluations by the Department and others that use inmate recidivism rates as performance measures.

For more information about recidivism research, see Lewis, John L., and Ensley, David T. (2002) "Recidivism" in Levinson, David, ed., Encyclopedia of Crime and Punishment, Vol. 3, Thousand Oaks, CA: Sage Publications, pp. 1352-1357.