Preliminary validation of the child abuse potential inventory in Chile
Loading...
Date
1998
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
PERGAMON-ELSEVIER SCIENCE LTD
Abstract
Objective: The objective of this work is a preliminary validation in Chile of the Child Abuse Potential (CAP) inventory (Milner, 1986).
Method: A sample of 134 participants was drawn from Santiago's Metropolitan Area and the city of Iquique in Chile, divided into two groups: people identified as physical abusers of their own children and people identified as nonabusers. As a result of the sampling procedure, the study group was composed of 64 women and three men. This distribution was matched in the control group. A reliability analysis, a predictive and construct validity analysis, and an item analysis of the Abuse Scale proposed by Milner were performed.
Results: The Abuse Scale items were tabulated both with the weighted scores proposed by Milner and with simple scores. The results were more consistent with simple scores. It correctly classified 90% of respondents (91% in the abuse group and 88% in the control group). Of the 76 items of the Abuse Scale, 55 were highly discriminatory, six of them were nonsignificant and 15 of them had differences that were not great enough so as to discriminate people well. To evaluate the dimensionality of the items, a factor analysis was carried out. The best solution (Oblimin rotation) was obtained with 6 factors, which accounts for 53.8% of the variance.
Conclusion: Milner's Abuse Scale discriminated between abusing and nonabusing individuals, and showed a factor analysis similar to the original one. Some items reflected cultural conducts in Chile instead of potential abuse: they are related with socially desirable nearness and cleanness of children and home, specially in low-income families. (C) 1998 Elsevier Science Ltd.
Method: A sample of 134 participants was drawn from Santiago's Metropolitan Area and the city of Iquique in Chile, divided into two groups: people identified as physical abusers of their own children and people identified as nonabusers. As a result of the sampling procedure, the study group was composed of 64 women and three men. This distribution was matched in the control group. A reliability analysis, a predictive and construct validity analysis, and an item analysis of the Abuse Scale proposed by Milner were performed.
Results: The Abuse Scale items were tabulated both with the weighted scores proposed by Milner and with simple scores. The results were more consistent with simple scores. It correctly classified 90% of respondents (91% in the abuse group and 88% in the control group). Of the 76 items of the Abuse Scale, 55 were highly discriminatory, six of them were nonsignificant and 15 of them had differences that were not great enough so as to discriminate people well. To evaluate the dimensionality of the items, a factor analysis was carried out. The best solution (Oblimin rotation) was obtained with 6 factors, which accounts for 53.8% of the variance.
Conclusion: Milner's Abuse Scale discriminated between abusing and nonabusing individuals, and showed a factor analysis similar to the original one. Some items reflected cultural conducts in Chile instead of potential abuse: they are related with socially desirable nearness and cleanness of children and home, specially in low-income families. (C) 1998 Elsevier Science Ltd.
Description
Keywords
child abuse screening, psychometric characteristics of instrument, transcultural validation of instrument, RISK, VALIDITY