PREDICTING PRISON MISCONDUCTS

A meta-analysis was conducted on 39 studies that generated 695 correla tions with prison misconducts. Predictors of prison misconducts were g rouped into 16 domains as follows: (a) personal characteristics (n = 9 ), (b) situational factors (n = 3), and (c) actuarial measures of anti social personality and risk (n = 4). Personal and situational variable s were similar in their ability to predict prison misconduct. Within t hese two categories, antisocial attitudes and behavior (e.g., companio ns, prison adjustment), criminal history, and institutional factors we re the strongest predictors. Among actuarial measures, an interview-ba sed risk protocol produced the highest correlations with prison miscon ducts. The prediction of violent misconducts was associated with great er effect sizes than nonviolent misconducts. Despite the limitations o f the database, several recommendations for assessing prison misconduc ts appear warranted.