Distinguishing teaching interactions of physically abusive from nonabusive parent-child dyads.

Researchers using a variety of behavioral coding instruments have demonstrated differences between the interactions of physically abusive parent-child dyads and those of nonabusive pairs. Despite fair consistency among these findings, there remains controversy regarding the value of observation of parent-child interaction to the clinical evaluation of child abuse. This study was an investigation of the ability of child protective service workers to distinguish videotaped interactions of physically abusive parent-child dyads from interactions of nonabusive dyads. The interactions occurred as part of a teaching task and were nondisciplinary. The child protection workers achieved a 76% rate of accuracy in identifying the abuse status of the dyads based on observation of only three minutes of semi-structured videotaped interaction. Of the workers, 40% were accurate in classifying 100% of the dyads observed. Increased protective service experience was not associated with increased accuracy, but subjects with no experience in protective service performed at chance level only. The results of this study show that experienced professionals can detect interactional differences in nondisciplinary parent-child interchanges of abusive as compared to nonabusive dyads. This suggests that clinical observation of parent-child interaction may be one important diagnostic tool within a comprehensive evaluation of families suspected of physical child abuse.