Neural network modeling of risk assessment in child protective services

The advantages of using neural network methodology for the modeling of comp lex social science data an demonstrated, and neural network analysis is app lied to Washington State Child Protective Services risk assessment data. Ne ural network modeling of the association between social worker overall asse ssment of risk and the 37 separate risk factors from the State of Washingto n Risk Assessment Matrix is shown to provide case classification results su perior to linear or logistic multiple regression. The improvement in case p rediction and classification accuracy is attributed to the superiority of n eural networks for modeling nonlinear relationships between interacting var iables; in this respect the mathematical framework of neural networks is a better approximation to the actual process of human decision making than li near, main effects regression. The implications of this modeling advantage For evaluating social science data within the framework of ecological theor ies are discussed.