U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

Assessing Intimate Partner Violence in Incarcerated Women

NCJ Number
211491
Journal
Forensic Nursing Volume: 1 Issue: 3 Dated: Fall 2005 Pages: 106-110
Author(s)
Michele J. Eliason; Janette Y. Taylor; Stephan Arndt
Date Published
2005
Length
5 pages
Annotation
This study assessed the psychometric qualities of the Index of Spouse Abuse (ISA), a screening instrument for intimate partner violence, for use with incarcerated women.
Abstract
The ISA is a 30-item scale designed to assess the existence of abuse as well as its severity. In the current study, the ISA was part of an instrument packet designed to obtain information about children of incarcerated mothers, including the presence of violence in the home. The packet was completed by 149 women housed in a prison-based therapeutic community drug treatment program. Using Hudson and McIntosh's two-factor solution and cut-off scores, 67 percent of the sample had experienced physical abuse, and 54 percent had experienced nonphysical abuse. Nearly 74 percent of the sample had experienced some type of abuse, and 46 percent exceeded the cut-off score for both types of abuse. The findings indicate that the ISA is a psychometrically sound instrument for use in research with incarcerated women and in their treatment planning. It can be rapidly and easily administered and reliably detects both the presence and severity of intimate partner abuse for women. The authors note, however, that the high alpha coefficients found overall in the study, as well as several other studies, indicates that the ISA could be simplified to produce one score. The ISA's weighted score system makes it difficult to use clinically, thus limiting its usefulness to the field. A computer-administered version with the weights added automatically would remove this limitation, as would a single-factor score. 33 references