A Comparison of Prospective and Retrospective Reports of Child Abuse

Helene Raskin White, Rutgers University
Ping Hsin-Chen, UMDNJ - New Jersey Medical School
Cathy Spatz Widom, New Jersey Medical School (UMDNJ)

Most studies of child abuse have not collected prospective data. Instead they have waited until adulthood to collect measures of abuse and neglect and have used these data to conduct analyses examining the effects of childhood victimization. The validity of the results from these retrospective studies is questionable and few prospective studies have assessed the accuracy of recall. In this study, we make use of unique data from a longitudinal study of adolescents and young adults to examine how well individuals can recollect parenting behavior, discipline practices, physical abuse and sexual experiences and abuse. Data were collected as part of the Rutgers Health and Human Development project, a prospective cohort-sequential, longitudinal study of adolescent development. The sample for the present analysis consists of 374 men and women who were interviewed five times between the ages of 12 years and 30/31 years. Overall, the data indicate weak reliability for recall of parenting behaviors and at best moderate reliability for recall of discipline practices and physical abuse. Reliability of recall of the timing of sexual experiences is high, although actual reports of sexual abuse have low reliability. Results from studies based on retrospective accounts of physical and sexual abuse should be interpreted with caution.

(Return to Program Resources)

Updated 05/20/2006