Aim: This study aimed to develop a detection scale for intimate partner violence (DS-IPV), excluding physical and sexual violence, and to validate the scale.
Methods: A cross-sectional survey was undertaken using an anonymous 26-item self-administered questionnaire with volunteer women aged 18-59 residing in one prefecture in Japan. A total of 1,937 responses were obtained, and 1,915 were valid. The responses of the 1,277 women who had a partner at the time were included in the analysis. Statistical analysis examined exploratory/confirmatory factor analysis, internal consistency, and construct validity.
Results: A four-factor scale with 22 items was extracted for the DS-IPV. In the confirmatory factor analysis, the indices of fitness supported these results. Prior to the factor analysis, the Kiser-Mayer-Olkin measurement (.936) and Bartlett's sphericity test (p < .001) showed that all samples met the statistical criteria. Cronbach's alpha coefficient was .92 for total score and varied between .70 and .90 for each of the four factors, and internal consistency was confirmed. The following four factors emerged from an oblique factor analysis, with a cumulative variance of 58.1%: "arousing anxiety"; "controlling behavior"; "coercion and threat"; and "daily emotions". These four factors had a moderate correlation (.23-.48, p < .01) with each other, which indicated construct validity. Confirmatory factor analysis showed the indices of fitness supported these results.
Conclusions: The DS-IPV scale, a concise scale to detect and screen for potential intimate partner violence, was developed. The DS-IPV scale showed good internal consistency and construct validity.
Keywords: detection scale development; female victim; intimate partner violence; reliability; validity.
© 2020 Japan Academy of Nursing Science.