In this paper and issue brief, we look at the history of the Child and Family Services Review (CFSR) process and the national standards that are central in that process. CFSR is directed at the goal of improving the performance of state child welfare systems, which is clearly desirable. However, the current national standards in the CFSR process do not optimally support that goal. We review the methods used to derive the national standards--six numerical targets that state child welfare systems are expected to meet. After an introductory overview of the CFSR process, we discuss the variation among states and how that affects the quality and use of the data reported to ACF for determining national standards. We then turn to the analytic approach that has been applied to these data; specifically we focus on the development of “composite scores” for four of the six standards.

The development of the national standards involved a complex analytic technique (principal components analysis) that was not necessary. Further, the national standards were set arbitrarily. Finally, the expectations for minimal improvement of states falling below the national standards are based on flawed statistical methods.

A functional, useful CFSR process would encourage improvement with a clear and coherent use of available data, with an eye to ways to improve both the quality and quantity of data related to children and families who come to the attention of the child welfare system.