OBJECTIVES: Endometriosis is a common, chronic gynaecological disease among women of reproductive age. Understanding treatment effects for endometriosis is critical for benefitting the health of women with this condition. The SF-36 has utility for understanding this condition because of some of its subscale focus (pain) and its ability to compare across diseases and populations. The SF-36 is one of the most common health measures used in studies of endometriosis; however, it has not been validated for this disease. The goal of this study was to assess validity of the SF-36 for endometriosis. METHODS: Using data from two clinical trials (N=252 & 198) of treatment for endometriosis, a full complement of psychometric analyses was performed. Additional instruments included a pelvic pain VAS and a clinical global impression of change (CGI-C). Analyses were conducted using Stata 10.1 and Mplus 5.1. RESULTS: Factor analyses confirmed the standard 8-subscale and 2-component factor structure of the SF-36 (CFI=0.94 and 0.93 for the two trials). The Bodily Pain (BP) subscale and the PCS were both moderately correlated with the pain VAS at baseline and over time (r = -0.37 to -0.62; p<0.001). Change in BP and PCS correlated with the CGI-C (r = -0.30 & -0.43; p<0.001): those who had the greatest change in BP and PCS also reported the greatest change on CGI-C. Other subscales (Role Physical, General Health, Vitality, Social Functioning) showed smaller correlations with change in the pain VAS and CGI-C. CONCLUSIONS: The SF-36 –particularly BP and the PCS – is sensitive to differences in patient pain and change in pain experience in endometriosis. The BP subscale and the PCS were dimensions of the SF-36 that showed consistent sensitivity to pain differences and change; however, other SF-36 subscales were significantly, but less strongly, related to differences in severity and changes in pain.