An MCMI-III Discriminant Function Analysis of Incarcerated Felons

Abstract

Discriminant function analyses were employed to predict institutional misconduct in a sample of 128 incarcerated offenders. Misconduct was divided into two categories: reprimands and behavioral consequences (i.e., early lockups, segregation, monetary penalties, and suspensions). Results indicated that the MCMI-III provides greater predictive utility than does demographic information alone. In the prediction of reprimands, an overall correct classification rate of 79% was obtained using both demographic and MCMI-III data. A correct classification rate of 75% was found in the prediction of misconduct related to more behavioral ramifications. The clinical implications of these results and directions for future research are discussed.