Moving child abuse and neglect prevention programs forward: improving program evaluations

This paper focuses on evaluations of programs to prevent child abuse and neglect. It discusses the quality of the evaluations and offers suggestions for improvement to evaluators, program planners, and funders. Thirteen evaluations recommended by experts and published from 1978 through 1988 were reviewed. Using the true experiment as the standard of quality, we found evaluations frequently characterized by careful attention to methodologic detail. Evaluators describe eligibility criteria; half of the studies had control groups or followed progress. However, to learn more about prevention of abuse and neglect the caliber of evaluation research must improve. Evaluators do not define abuse or neglect; they accept reports of abuse. This has the effect of shifting the responsibility to others whose concerns and reporting standards differ. Lack of definition is reflected in a paucity of valid measurements; we could not assess the characteristics of families who benefit most from programs. Evaluators also omit important topics. Although seven evaluations studied health-related variables, no comprehensive analysis of the consequences and costs of medical neglect could be found; very little information on costs and benefits of any kind was available.