The discussions emphasize that relatively little is known about how to prevent or treat spouse and child abuse and that research on its incidence and etiology and on effective means of intervention are essential to efforts to protect its victims. They also note that differences in definition and methodology produce disparate estimates of the incidence of abuse, that no single cause exists for family violence. Therefore, research terms should be standardized, the mixture of social and individual factors that lead to violence should be identified, and a multifaceted approach rather than a single "cure" should be emphasized in treating family violence. In addition, programs evaluations should be designed to reveal service inadequacies, including points of delay, staff inadequacies, decisionmaking problems, and administrative weaknesses that result in unnecessary injuries after violence has been reported. Tables and reference notes.