As a special group of police officer, prison police have to endure more work stress and have significant work-family conflict, which may lead to more physical and mental health problems and need to be noticed by the society. The Work-Family Conflict Scale (WFCS) is a brief self-report scale that measures the conflict that an individual experiences between their work and family roles and the extent they interfere with one another. However, there is limited data on the scale's psychometric properties. The aim of this study was to examine the dimensionality and reliability of the Chinese version of the WFCS. The study sample was made up of a total of 717 Chinese prison police (64.7% male, M = 41.73 years, SD = 8.30 years). The Rasch Rating Scale Model (RSM) was used to determine the latent structure and estimate the quality of items and reliability of scale. The principle component analysis (PCA) showed that the assumption of unidimensionality was fulfilled. The infit and outfit mean square (MNSQ) statistics (0.84-1.47) were of a reasonable range, and point-measure correlations (0.64-0.79) indicted good model fit of each item. The item-person separation and reliability indices both met psychometric standards, illustrating good reliability. The person-item map indicated acceptable fit of items and persons, suggesting an alignment between persons and items. In addition, no evidence emerged of differential item functioning across different gender groups. Overall, the WFCS has good reliability and validity, and can be used to accurately evaluate the level of work-family conflict in Chinese prison police.