Is there an allegiance effect for assessment instruments? Actuarial risk assessment as an exemplar.

Abstract

The Violence Risk Appraisal Guide (Harris, Rice, & Quinsey, 1993), the Sex Offender Risk Appraisal Guide (Quinsey, Harris, Rice, & Cormier, 1998), and the Static-99 (Hanson & Thornton, 1999) are three well-researched risk measures. All three instruments were good predictors of recidivism (r = .31). However, these effect sizes were significantly larger in studies conducted by the instruments’ authors (r = .37) than in studies conducted by independent researchers (r = .28). This allegiance effect remained significant even when the initial validation studies were excluded. No other design or sample characteristics were significant moderators of the relation between scores and recidivism. These findings raise questions about whether such an allegiance effect may be found for other measures.