Prediction accuracy of the Washington and Illinois risk assessment instruments: An application of receiver operating characteristic curve analysis

Abstract

In recent years many child protective services (CPS) agencies have used structured risk assessment to streamline and rationalize case identification and management in a resource-depleted environment. Empirical evidence of risk assessment's capacity to predict child maltreatment has been limited to tests of single instruments. This article compares the predictive performances of the Illinois CANTS 17B and the Washington State Risk Matrix on a sample of New Jersey CPS cases using logistic regression and receiver operating characteristic (ROC) curve analysis. Both instruments predict case recidivism, closings, and substantiation with probabilities greater than chance. Moreover, in the prediction of case recidivism, the Washington instrument showed superiority at a statistically significant level. The implications of ROC analysis for child welfare decision makers and the relationship among test accuracy, problem prevalence, and diagnostic cutoff points are discussed.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 51 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.