Base Rate Information Improves Risk Estimation of Laypeople and Experts

Interpretation of psychological assessments using actuarial risk tools varies according to risk level and communication format. However, previous research in this area has widely neglected the fact that decision makers' probability judgments often exceed empirically observed recidivism rates derived from follow-up studies. Thus, the current studies investigated the additional effect of base rate information on the estimation of risk including laypeople (n = 283, 68% female, M = 24 years old) and expert witnesses (n = 62, 76% female, M = 42 years old). Results largely replicate findings of previous research regarding the effects of risk level and communication format. Both groups provided estimates substantially higher than expected official recidivism rates (laypeople: M = 56.7%, experts: M = 28.1% vs. empirical base rate of 10%). Providing base rates reduced this discrepancy by 12% points (Cohen's d = 0.43) for laypeople and 7% points (d = 0.42) for experts. Additional analyses indicated that experts specializing in risk assessment produced estimates closer to follow-up benchmarks than those focusing on other forensic tasks. Years of professional experience among expert witnesses were positively associated with larger discrepancies between estimates and follow-up rates. These findings underscore the importance of incorporating base rate information to anchor risk judgments in empirically established benchmarks. Communication format had, at most, a minimal impact on risk estimates, whereas providing base rate information meaningfully reduced discrepancies between estimates and follow-up benchmarks in both lay and expert samples. We recommend reporting empirically grounded base rates alongside any chosen communication format.