Echeburúa Enrique, Amor Pedro Javier, Loinaz Ismael, de Corral Paz
Universidad del País Vasco, Facultad de Psicología, San Sebastián, Spain.
Psicothema. 2010 Nov;22(4):1054-60.
The aim of this study was to describe the psychometric properties of the Severe Intimate Partner Violence Risk Prediction Scale and to revise it in order to ponderate the 20 items according to their discriminant capacity and to solve the missing item problem. The sample for this study consisted of 450 male batterers who were reported to the police station. The victims were classified as high-risk (18.2%), moderate-risk (45.8%) and low-risk (36%), depending on the cutoff scores in the original scale. Internal consistency (Cronbach's alpha=.72) and interrater reliability (r=.73) were acceptable. The point biserial correlation coefficient between each item and the corrected total score of the 20-item scale was calculated to determine the most discriminative items, which were associated with the context of intimate partner violence in the last month, with the male batterer's profile and with the victim's vulnerability. A revised scale (EPV-R) with new cutoff scores and indications on how to deal with the missing items were proposed in accordance with these results. This easy-to-use tool appears to be suitable to the requirements of criminal justice professionals and is intended for use in safety planning. Implications of these results for further research are discussed.
本研究的目的是描述《严重亲密伴侣暴力风险预测量表》的心理测量特性，并对其进行修订，以便根据20个条目的区分能力进行加权，并解决缺失条目问题。本研究的样本包括450名向警察局报案的男性施暴者。根据原始量表的临界分数，受害者被分为高风险（18.2%）、中度风险（45.8%）和低风险（36%）。内部一致性（克朗巴哈系数α = 0.72）和评分者间信度（r = 0.73）可以接受。计算每个条目与20条目量表校正总分之间的点二列相关系数，以确定最具区分性的条目，这些条目与上个月亲密伴侣暴力的背景、男性施暴者的特征以及受害者的脆弱性有关。根据这些结果，提出了一个修订版量表（EPV-R），其中有新的临界分数以及关于如何处理缺失条目的说明。这个易于使用的工具似乎符合刑事司法专业人员的要求，旨在用于安全规划。讨论了这些结果对进一步研究的意义。
