Standardized Tests for Inmates

Standardized assessments are not just for school anymore. More than fifteen states have created policies that now require correction facilities to implement modern risk-assessment methods when deciding the fates of prisoners. Some states, including Ohio, which I will discuss later, have begun using computerized assessments. In Michigan, officials state that the assessments have helped reduced the state’s prison population “by more than 15% from its peak in 2007 and with lowering the three-year recidivism rate by 10 percentage points since 2005”. These statistics are quite compelling, but we should not overlook other factors that have contributed to the decreasing recidivism rate including transfer of inmates from state to local jails, and treatment given during incarceration.

figure from "State Parole Boards Use Software to Decide Which Inmates to Release" Wall Street Journal 

I am not sure why these charts were included in the article. Unfortunately these charts present the Ohio Risk Assessment System (ORAS) as an overly simplistic tool when in fact I have learned that ORAS is quite complex and more thorough than presented.

The first graph shows a decrease of prisoners returning to prison, with the biggest decrease beginning in 2008. However, we are not informed as to what contributed to this big decrease. Moreover, ORAS was not introduced until 2011 so there must be other factors that have helped decrease recidivism other than computerized testing. We could better gauge the effectiveness of ORAS if the graph was extended another five years.

I found the second chart troubling, as I don’t understand how these questions would assess an inmate’s probability of returning to jail after being released, especially if they are all equally weighted. Isn’t it easy to lie about whether you believe it’s possible to overcome the past? If I were trying to get out of prison, I would readily answer yes. After more research, I learned that the ORAS is actually composed of seven assessment tools to assess the risk factors of inmates during their time in prison (Ohio.gov). Each assessment tool consists of several detailed items (i.e. past behavior, family history) that are given points. Some items however have “multiple increasing risk scores, and as a result were scored with increasing values” (Latessa et al 18).  In other words, lying doesn’t necessarily mean freedom.

The third chart is a mathematical wonder. First, it is interesting how the percentages of males add up to 135% and for females, 106.5%. If all the groups don’t add up to one overall sample, then each group then must be its own separate sample. Nope! The low-risk group adds up to 27.5%, the moderate-risk group at 94% and high-risk group at 120%. So exactly what are these percentages referring to? I tried looking for the answer myself in an article written by the University of Cincinnati, but was not able to find data that matched the one in the article. Where did this article get these percentages??

Lastly, like other standardized tests, it's important that we see assessments, like the ORAS, as tools and not true reflections of those they assess.

Sources

Walker, Joseph. "State Parole Boards Use Software to Decide Which Inmates to Release." Wall Street Journal 11 Oct. 2013. Web. 23 Oct. 2013. http://online.wsj.com/news/articles/SB10001424052702304626104579121251595240852

Latessa, Edward J., et al. The Creation and Validation of the Ohio Risk Assessment System (ORAS).  Ohio, 2008. Print.

Ohio.gov. “Ohio Risk Assessment System.” Ohio Department of Rehabilitation and Correction. 29 Jan 2013. Web. 23 Oct. 2013. http://www.drc.ohio.gov/web/oras.htm