Variables affecting psychiatrists' and teachers' assessments of the dangerousness of mentally ill offenders.

Nine high school teachers and four forensic psychiatrists rated the psychiatric assessments, histories, and offense descriptions of 9 child molesters, 10 property offenders, and 11 serious offenders against adults. The three types of patient data were rated separately and together on the likelihood of a property offense, the likelihood of an assaultive offense, and the seriousness of an assault should one occur, if these male patients were to be released from a maximum security mental hospital. Raters also judged whether the patients should be released. There was very low interrater congruence in all ratings and in the release decisions for both occupational groups, although the correlations between the occupational group averages were very high. Ratings of the combined information were well predicted from a linear combination of the ratings of its three components for both rater groups. Psychiatric assessment data made little contribution to the appraisal of the overall file. These data question the expertise of forensic psychiatrists as well as the usefulness of psychiatric assessment data in the prediction of dangerousness.