NCJ Number
55733
Date Published
1978
Length
78 pages
Annotation
USE OF VIOLENCE PREDICTION IN THE CRIMINAL JUSTICE AND MENTAL HEALTH SYSTEMS, THE STATE OF RESEARCH AND PREDICTION ACCURACY, AND RECOMMENDED RESEARCH DIRECTIONS ARE DISCUSSED.
Abstract
THE TASK OF IDENTIFYING VIOLENCE-PRONE INDIVIDUALS HAS BEEN ALLOCATED TO THE CRIMINAL JUSTICE AND MENTAL HEALTH SYSTEMS. BUT BOTH SYSTEMS USE THESE PREDICTIONS AS VARIABLES IN DECIDING WHO SHOULD BE INSTITUTIONALIZED. THE GRAVITY OF THIS SITUATION MAKES SEVERAL RESEARCH ISSUES SIGNIFICANT: (1) THE ABILITY TO PREDICT WHO WILL ENGAGE IN VIOLENT BEHAVIOR IS VERY POOR, BUT IT IS POSSIBLE TO IDENTIFY PERSONS WHO ARE AT HIGH RISK; (2) THE BEST PREDICTOR OF FUTURE VIOLENT BEHAVIOR IS A RECORD OF PAST VIOLENT BEHAVIOR, WHILE THE POOREST PREDICTORS ARE THOSE THAT RELATE TO PSYCHOLOGICAL FUNCTIONING; (3) ACTUARIAL TABLES MAY BE SUPERIOR TO CLINICAL JUDGMENTS IN PREDICTING VIOLENT BEHAVIOR; AND (4) CLINICAL PREDICTION ALLOWS SOCIALLY SENSITIVE PREDICTOR VARIABLES TO BE HIDDEN. PREDICTIONS ARE INACCURATE FOR SEVERAL REASONS, INCLUDING LACK OF CORRECTIVE FEEDBACK, DIFFERENTIAL CONSEQUENCES TO THE PREDICTOR, DIFFERENTIAL CONSEQUENCES TO THE SUBJECT, ILLUSORY CORRELATION, UNRELIABILITY OF THE CRITERION, AND POWERLESSNESS OF THE SUBJECT. IT IS RECOMMENDED THAT RESEARCH ON VIOLENCE PREDICTION EMPLOY MULTIPLE DEFINITIONS OF VIOLENCE, MULTIPLE TIME PERIODS FOR FOLLOWUP VALIDATION, AND MULTIPLE METHODS OF VERIFYING THE OCCURRENCE OF VIOLENT BEHAVIOR. FUTURE RESEARCH PRIORITIES SHOULD FOCUS ON VIOLENT BEHAVIOR AND UNEMPLOYMENT, VIOLENT BEHAVIOR AND CIVIL COMMITMENT, AND CORPORATE VIOLENCE. ACKNOWLEDGEMENTS, REFERENCES, SOME TABULAR DATA INCLUDED. (LWM)