Development and Validation of the Intimate Partner Violence Scale-5 (IPVS-5): Psychometric Properties in a Large-Scale Chinese Study

Background: Intimate partner violence (IPV) poses serious threats to both physical and mental health and is increasingly recognized as a major public health concern. Despite its prevalence in China, culturally sensitive and psychometrically robust IPV scales remain scarce. This study aimed to develop and validate a brief scale for IPV.Methods: Data were obtained from the Psychology and Behavior Investigation of Chinese Residents (PBICR 2023), which included 23,469 participants. The scale development process involved a literature review and expert consultations. Psychometric evaluation included item analysis, exploratory and confirmatory factor analyses, internal consistency assessment, and tests of convergent validity with depression (PHQ-9), anxiety (GAD-7), and stress (PSS-4). Latent profile analysis (LPA) and receiver operating characteristic (ROC) analysis were also conducted.Results: The IPVS-5 demonstrated excellent internal consistency (Cronbach’s α = 0.947), a clear unidimensional factor structure, and good convergent validity (r = 0.45–0.47 with PHQ-9, GAD-7, and PSS-4). LPA revealed three distinct IPV risk profiles: low, moderate, and high. The ROC analysis showed excellent discriminative ability, with an area under the curve (AUC) of 0.999. Optimal integer cut-off scores of >8 (indicating moderate-to-severe IPV) and >13 (indicating severe IPV) were used to categorize IPV severity into interpretable ranges: Low (5-8), Moderate (9-13), and High (14-15).Conclusion: The IPVS-5 is a valid, reliable, and efficient scale for identifying IPV. It holds significant potential for use in clinical, public health, and community settings to support early detection and intervention.