MOTHER-CHILD INTERACTIVE PATTERNS IN ABUSIVE FAMILIES VERSUS NONABUSIVE FAMILIES - AN OBSERVATIONAL STUDY

Citation
Ma. Cerezo et al., MOTHER-CHILD INTERACTIVE PATTERNS IN ABUSIVE FAMILIES VERSUS NONABUSIVE FAMILIES - AN OBSERVATIONAL STUDY, Child abuse & neglect, 20(7), 1996, pp. 573-587
Citations number
47
Categorie Soggetti
Social Work
Journal title
ISSN journal
01452134
Volume
20
Issue
7
Year of publication
1996
Pages
573 - 587
Database
ISI
SICI code
0145-2134(1996)20:7<573:MIPIAF>2.0.ZU;2-C
Abstract
The interest in aspects relative to the maintenance of aversive family interactions is common among investigators who, from many different f ronts, treat child and family issues. If these interactions end in epi sodes of physical abuse of a child by his/her parents, the need to qui ckly detect these patterns of family violence in order to prevent and treat them is all the more urgent. The present work falls within this context and focuses on a detailed analysis of behavior in the home of both the abusive mother and the victimized child. In line with recent investigations, the work also centers on the analysis of the actual mo ther-child interaction by carrying out a focused analysis of the behav ior of the dyad given that these behaviors greatly affect each other. The study was conducted with a group of 23 mother-child dyads that pre sented episodes of physical abuse towards the child, and a control gro up of 24 dyads that did not present any interactional problems. The dy ads were observed at home by a group of trained observers using the SO C III system of direct observation (Standardized Observation Codes III rd revision, Cerezo, 1991; Cerezo, Keesler, Dunn, & Wahler, 1986). Thr ee stepwise discriminant function analyses were carried out for the va riables relative to the maternal behavior, the child's behavior, and t he measurement of the mother-child sequence of interaction. The result s indicate that behaviors with a negative valence better classify the subjects into the groups to which they belong, obtaining a better clas sification when the variables relative to the interaction are used in the analysis. It can be concluded that it is necessary to include this measurement in future analysis due to its great classificatory and pr edictive value.