Background: There has not yet been a pictorial version of a patient-reported outcome measure for shoulder pain. Purpose: To translate the English version of the Oxford Shoulder Score (OSS) to a simplified Chinese version (SC-OSS) and to validate a new face-scale version of the OSS (FS-OSS), while investigating cross-cultural adaptation, validation, and reproducibility of both versions in patients with shoulder pain. Study Design: Cohort study (diagnosis); Level of evidence, 2. Methods: The translation and cross-cultural adaptation of the SC-OSS was performed using a forward-backward translation method. The FS-OSS was developed on the basis of the SC-OSS, using the Wong-Baker FACES Pain Rating Scale for reference. Participants were asked to complete the SC-OSS, FS-OSS, Simple Shoulder Test (SST), Constant-Murley score (CMS), and 36-Item Short Form Health Survey (SF-36). Validation and reproducibility were tested by calculating Cronbach α values for internal consistency as well as by intraclass correlation coefficients. Time needed to complete the scores was used to test cross-cultural adaption. Results: A total of 312 respondents participated in the research and completed all outcome measures. The internal consistency was strong, with a Cronbach α of .94 and .91 for the FS-OSS and SC-OSS, respectively. High intraclass correlation coefficient values for the FS-OSS score (0.95) and SC-OSS (0.92) were obtained, which indicated excellent test-retest reliability. The Pearson correlation coefficients of the SC-OSS and FS-OSS with the SST ( r = 0.67 and 0.65, respectively), CMS ( r = 0.62 and 0.66, respectively), and SF-36 ( r = 0.52 and 0.57, respectively) indicated good construct validity. The time needed to complete the FS-OSS was less than that needed for the SC-OSS and SST. Conclusion: The FS-OSS and SC-OSS were validated as reliable instruments for patients with shoulder pain. For Chinese patients, the face-scale version was easier to understand than the cross-cultural text version.