Criminal Recidivism Study Falls Short


Recently a long awaited study was published on the recidivism rates of those convicted of certain criminal activity. The “Blumstein Study” was conducted by Carnegie Mellon professor Alfred Blumstein and co-author Kiminori Nakamura. They examined New York State Arrest records for more than 88,000 individuals’ ages 16 to 20 who were first arrested in 1980 for robbery, burglary or aggravated assault. They then analyzed recidivism rates for these individuals over the next for 27 years through 2007. They concluded that ex-offenders do in fact have a high rate of recidivism within the first five years after arrest. However, the study suggests that for offenders who “stay clean” for five years or more are much less likely to get arrested. They conclude that those who have not committed additional offenses after five years are almost as likely to commit a crime as the rest of the general population of the United States.

The study is timely and important for many reasons. First, the Obama administration has made it a point to break down barriers for individuals who have committed crimes to gain employment. Secondly, the EEOC has a mission to ensure a disparate class of ex-offenders is not created in the United States. Combine that with their contention that minorities are arrested and prosecuted at a much higher rate than Caucasians. The EEOC’s E-RACE (Eradicating Racism and Colorism from Employment) initiative is designed to improve their efforts to ensure workplaces are free of race and color discrimination. According to their website:

Specifically, the EEOC will identify issues, criteria and barriers that contribute to race and color discrimination, explore strategies to improve the administrative processing and the litigation of race and color discrimination claims, and enhance public awareness of race and color discrimination in employment. As a framework for implementing the E-RACE Initiative, EEOC has developed a set of detailed E-RACE goals and objectives to be achieved within a 5-year time frame from FY 2008 to FY 2013.

Additionally, the Commission will combine the objectives of E-RACE with existing EEOC initiatives. For example, the Commission will integrate the goals of the Systemic Initiative by addressing race and color issues with class and systemic implications. It will incorporate the principles of the Youth@Work Initiative by combating disparate treatment of youth based on race and color. And, the Commission will complement the outreach and enforcement efforts of the LEAD Initiative by challenging exclusionary employment policies that adversely impact people of color who also have disabilities (in both the private and public sectors).

Finally, the Commission will strengthen partnerships with employee advocates and state and local human rights commissions and increase its outreach to human resource professionals and employer groups to address race and color discrimination in the workplace.

The underlying issue is that the Blumstein study has a lot of holes. Noted employment screening industry experts and members of the National Association of Professional Background Screeners (NAPBS) have commented on this study for months. Recently, Security Management Magazine published a great article on the topic as well. One well respected expert concluded that because the study was limited to New York State it fails to find those criminals who may have committed crimes in other states. Blumstein acknowledges this and expects that might lead to a 10 percent increase in the risk findings after that data is added. The same expert also notes that because the study only looks at arrests and not convictions the true sample size is greatly reduced.

In my opinion these industry experts are spot on! Having examined the findings of the study I find even more shortcomings. First, what are the parallels between “major” and “minor” crimes; meaning, how does the recidivism rate for someone convicted of armed robbery compare to someone convicted of petty theft? Secondly, the study is only able to track the rate of those who committed the same crime a second time and only for those who were actually caught. Just because a person hasn’t been arrested or prosecuted doesn’t mean that they have been rehabilitated. It can mean that they just haven’t been caught. Lastly this study only analyzes those who committed robbery, burglary or aggravated assault. It excludes all other criminal activity, so at best; this study would only be valid for those ages 16 to 20 who committed those crimes.

Many speculate that employers should take special note of the study’s conclusion. Some even suggest that this information might be helpful to states focusing early prison release programs. Right now, in my opinion, nothing should be concluded from this study other than more research needs to be performed. While future studies might indeed create a clear picture, this study is simply a stepping stone for additional research that incorporates a greater spectrum of criminal activity, a larger sampling size, a more diverse age group and a nationwide analysis. I also think that rather than analyzing arrest records which often do not lead to convictions or to convictions of lesser offenses; they should focus only on convictions. Of course, we’ll be following future studies for more conclusive information.

Jason Morris is the president and C.O.O. of Cleveland-based EmployeeScreenIQ , a best practices provider of employment screening services throughout the U.S. and worldwide. Jason can be reached at (800) 235-3954 ext. 424 or