The studies were selected from several computerized databases. Only one of the studies assessed physical abuse. The analysis revealed that responses rates declined significantly over time, from an average of 68 percent before 1985 to 49 percent for more recent surveys. In addition, the unadjusted estimates of the prevalent experience among women and men of childhood sexual abuse were 22.3 percent and 8.5 percent, respectively. Moreover, study response rates and operational definitions of child abuse together accounted for half the estimates of the observed prevalence of child sexual abuse. The prevalence estimates adjusted for response rates were 16.8 percent for females and 7.9 percent for males; they were 14.5 percent and 7.2 percent when adjusted for operational definitions and excluding the broadest, noncontact category of child sexual abuse. Furthermore, the prevalence of child sexual abuse did not vary significantly over the three decades reviewed after adjustment for response rates and definitions. Findings indicate the need for a large, methodological rigorous, population-based study of child abuse to determine the prevalence of child abuse and guide effective prevention and treatment interventions. Table and 45 references (Author abstract modified)