Relationship between childhood abuse and assaultive behavior in adolescent male psychiatric patients.

Abstract

To test in part Berkowitz's hypothesis that parents who attempt to influence their children's behavior by frequent punishment appear to foster aggressiveness in their offspring . (1973, p. 114) 15 male adolescents (Group I ) were randomly selected from a state psychiatric institution. They had been admitted with a presenting problem of assaultive behavior and for these persons there was recorded evidence of at least two instances of assault while institutionalized. Group I1 comprised another randomly drawn group of 15 male adolescents from the same facility, but these showed no presenting problem of assaultive behavior and no reported incidence of assault while institutionalized. There were no apparent mean differences between the groups on length of institutionalization, presenting problems, age, and family background. The sample ranged in age from 13 to 18 yr. with approximately 65% males. Only the lower and lower middle socioeconomic levels were represented in the sample and population. Mean length of institutionalization was approximately 8 to 12 mo. with a range from 4 mo. to 245 yr. Medical records at the hospital included reports from all previous placements, treacments, and agencies; these were consulted to determine if the two groups differed on documented abuse by a caretaker, i.e., reported and verified by two sources. Abuse included physical, sexual, or severe deprivation. Evidence of abuse was found in medical records of 8 of the 15 children in Group I. There were six cases of physical abuse, one of sexual abuse, and one of physical abuse plus severe deprivation (physical injury and excessive deprivation of food). By contrast, in Group I1 evidence of abuse was found in the medical records of three of the 15 children. Two of these cases involved physical abuse and one sexual abuse. Considering the stringent criterion of verifying abuse through two sources, it is surprising that chi squared of 3.57 ( p < l o ) even approached significance. No doubt a less stringent criterion of documented abuse might have suggested a significant difference. (Unfortunately it was not possible to verify this.) Further invest~gatjon in other similar facilities is needed to determine if this trend is idiosyncratic to r h ~ s facility.