Social:Standard Predictor

Standard Predictor of Violence Potential

The Standard Predictor (SP) is a set of measures of risk for abuse and violence developed in 2010 by Robert John Zagar and William M. Grove on a sample of 1,127 youth and 1,595 adults. These tests were validated on a sample of 236 teens and adults with 120 different occupations by Robert John Zagar in 2013. The Standard Predictor (SP) is a self-report set of true and false and multiple choice questions that can be administered to either teens or adults. The Standard Predictor (SP) correlates well with the Wechsler Intelligence Scale, the Vineland Social Maturity Scale, the Bender Visual Motor Gestalt Test, the Gates-MacGinitie Comprehension, Speed and Accuracy, and Vocabulary Tests and the Stanford Achievement Test. The Standard Predictor (SP) raw scores translate into a percentile of risk for abuse and violence for either an adolescent or an adult.

According to Loyola University of Chicago Maude C. Clark Professor of Humanistic Psychology and former Director Life Development Center, Cornell University, James Garbarino “Several tests were chosen after an exhaustive review of the research on the sensitivity and specificity of actuarial evaluations for “return to court” or fitness for duty uses. Tests with high reliability and validity, simplicity, and relatively short time of administration were chosen…The Standard Predictor has the best sensitivity and specificity of any instrument measuring violence potential. Standard Predictor (SP) is an assessment of violence potential for adults with 98 in true-false or multiple choice format items.

The Standard Predictor for Adolescents (SP) has 139 items. The SP evaluates specific, historical self-descriptions and requires 15 minutes to complete. The SP has no items from any of the other tests and is a free-standing instrument with 98 for adults or 139 independent items distinct from other tests. This measure was successful in discriminating randomly selected violent offenders (1,595 adults and 1,127 adolescents) from matched controls with 97% success in a combined sample of 2,722. This is noteworthy because most tests in the literature attempting to predict criminal recidivism or “return to court” have success rates from 70%-80%.” “The model identifies 97% of those who end up as killers (missing only 3%), and it misidentifies as killers only 3% of those who do not end up killing anyone. Thus, in statistical terms, this approach offers excellent “sensitivity” (also called the “true positive rate” because it represents the portion of the actual positives that the model correctly identified as such-i.e., the 97% of the killers who were correctly classified as such). And it offers excellent specificity (also called the “true negative rate”-the 97% of the non-offenders who are correctly classified as such). It is extremely hard to achieve such high rates of sensitivity and specificity, at the same time, so this is a major achievement.

V-T-E Psychology | V-T-E Mental and behavioral disorders (F290-319) | V-T-E Psychiatric assessment (forensic assessment)

References

1. Zagar, Robert. J., & Grove, William. M. "Violence risk appraisal of male and female youth, adults, and individuals." Psychological Reports, doi: 107, 3, 983-1009.
2. Zagar, Robert. J., Joseph W. Kovach., Basile, Brother., Grove, W. M., John Russell Hughes, Kenneth G. Busch, Michael Zablocki, William Osnowitz, Jonas Neuhengen, Yutong Liu, & Zagar, Agatha. K. (2013). "Finding workers, offenders or students most at-risk for violence: actuarial tests save lives and resources. Psychological Reports." 113, 685-716.
3. Garbarino, James. (2015) "Listening to killers; lessons learned from my 20 years as a psychological expert witness." Oakland California: University of San Francisco Press.