Comparing Effect Sizes in Follow-Up Studies: ROC Area, Cohen's d, and r

By M. Rice, Grant T. Harris. Law and Human Behavior, 2005

1,638 citations recorded by the source corpus.

Subfields: Clinical Psychology, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/4f0fd14f2dcff9b1b426bd1e5ace93f7caf7cc99

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Clinical Psychology

1. J. Wormith
2. Michael T. Baglivio
3. R. K. Hanson
4. Aldert Vrij
5. Jeremy Coid
6. Min Yang
7. S. Wormith
8. L. Helmus

In Sociology and Political Science

1. J. Wormith
2. Michael T. Baglivio
3. R. K. Hanson
4. Jeremy Coid
5. Min Yang
6. D. Thornton
7. S. Wormith
8. L. Helmus

Citations in the corpus

• Assessing Risk for Sexual Recidivism: Some Proposals on the Nature of Psychologically Meaningful Risk Factors (2010, background): “Both d and AUC are based on similar statistical models (Swets, 1986), and the transformations between these metrics are well documented: for example, d of 0.20 = AUC of 0.556, d of 0.50 = AUC of 0.639, d of 0.80 = AUC of 0.714; Rice & Harris, 2005).”
• Eliciting cues to deception and truth: What matters are the questions asked (2012, background): “He gave as an example the difference in height between 15- and 16-year-old girls (Cohen, 1988; Rice & Harris, 2005).”
• Examining the Validity of a Juvenile Offending Risk Assessment Instrument Across Gender and Race/Ethnicity (2013, background): “50s is fairly modest and considered ‘‘small’’ in magnitude of effect size (Rice and Harris, 2005).”
• The accuracy of recidivism risk assessments for sexual offenders: a meta-analysis of 118 prediction studies. (2009, background): “Although the effect sizes for most of the measures were moderate to large by conventional standards (Cohen, 1988; Rice & Harris, 2005), there are no direct procedures for translating the observed effect sizes into statements concerning the normative value or utility of specific measures (e.g., is a…”
• A Comparison of Risk Assessment Instruments in Juvenile Justice (2013, background): “Attempts to standardize interpretations of the AUC vary (Royston, Moons, Altman, & Vergouwe, 2009; Tape, n.d.) and must take the base outcome rate into account (Rice & Harris, 2005).”
• Trait Repetitive Negative Thinking: A Brief Transdiagnostic Assessment (2014, methodology): “Receiver-operator characteristic analyses using RTQ-10 total scores resulted in a large (Rice & Harris, 2005) area under the curve (.86, SE =.01) when classifying individuals as belonging to either the undergraduate or clinical sample.”
• Do We Know Who Will Drop Out?: A Review of the Predictors of Dropping out of High School: Precision, Sensitivity, and Specificity (2013, methodology): “While outside the scope of the present study, we encourage future research in this domain to examine the use of continuous indicator outcomes, such as ROC “Area Under the Curve” (AUC) analysis (Swets, 1988), which is amenable to effect size comparisons such as meta-analysis (Rice & Harris, 2005).”
• False Positives , False Negatives , and False Analyses : A Rejoinder to “ Machine Bias : There ’ s Software Used Across the Country to Predict Future Criminals . And It ’ s Biased Against Blacks . ” (2017, neutral): “…as a standard measure in assessing diagnostic accuracy of risk assessments and has properties that make it not affected by base rate or sample size (Rice & Harris, 2005). percent of low-risk defendants were rearrested (White and Black respectively), 53-56 percent of medium-risk defendants were…”
• False Positives, False Negatives, and False Analyses: A Rejoinder to "Machine Bias: There's Software Used across the Country to Predict Future Criminals. and It's Biased against Blacks" (2016, neutral): “…as a standard measure in assessing diagnostic accuracy of risk assessments and has properties that make it not affected by base rate or sample size (Rice & Harris, 2005). percent of low-risk defendants were rearrested (White and Black respectively), 53-56 percent of medium-risk defendants were…”
• The efficacy of violence prediction: a meta-analytic comparison of nine risk assessment tools. (2010, methodology): “For eight studies that reported only the AUC values, a direct conversion of the AUC value to the d value was carried out on the basis of the table of M. E. Rice and Harris (2005).”