Disproportionality at the "front end" of the child welfare services system: an analysis of rates of referrals, "hits", "misses", and "false alarms".

Data from NIS-4, NCANDS, and the State of California were used to analyze the front end of the child welfare services system--the referral and substantiation components--in terms of the system's ability to diagnose or detect instances of child maltreatment. The analyses show that Blacks are disproportionately represented in rates of referral into the system. Moreover, the analyses demonstrate that the system is less accurate for Blacks than for other racial or ethnic groups. There is a higher rate of false positives (or "false alarms") for Blacks than for other groups--that is, referrals leading to unsubstantiated findings. There is also a higher rate of false negatives (or "misses") for Blacks than for other groups--that is, children for whom no referral was made but who are in fact neglected or abused. The rate of true positives (or "hits")--children for whom a referral has been made and for whom that allegation has been substantiated--is generally higher for Blacks than for other groups, but this is attributable largely to the higher rate of referral for Blacks. In sum, the system demonstrates lower levels of accuracy for Blacks than for other groups. A model is proposed demonstrating that random error, as opposed to systematic bias, could produce a pattern of results much like that observed in the data.