"The authority to detain a particular defendant to protect society is naturally premised on the idea that it is at least reasonably possible for the courts to make predictions regarding the possibility that particular defendants, if released, would act in ways that would endanger other individuals or the community in general” [D. N. Adair].

Over time, there has been a need for professionals to assess the level of dangerousness a certain individual poses to society.  This may take the form of assessing an individual for his tendency to commit violent acts, either now or in the future, or it may be assessing an inmate's suitability for release back into the community; that is, whether or not he will continue to pose a danger upon release.

Given that an inaccurate prediction could mean the difference between liberation and incarceration, these assessments must be conducted with great caution.  This is especially true when one considers that the rate of false positives, that is, those assessments that incorrectly identify an individual as a danger when he is not, is incredibly high. 

Predictions of dangerousness are made in several ways.  Predictions may be clinically based, where a professional will interview an individual in a controlled setting, or they may be based on statistical assumptions.

This article explores the issue of dangerousness and the broad ways in which predictions are made. First, the issues on which the prediction of dangerousness is based; then a look at the two very broad ways such predictions are made in the clinical setting and through the use of statistics and averaged offender types.  Next, I will address the problems with prediction and the accuracy of such predictions.  Finally, I will briefly address the situation of people who are mentally disordered.