This study was designed to assess the reliability and validity of a Pain Behavior Observation method with fibromyalgia syndrome (FS) subjects and to determine the factors which predict pain behavior among FS subjects. Fifty-eight female FS subjects participated in the videotaped Pain Behavior Observation method. Subjects also completed the McGill Pain Questionnaire (MPQ), the Arthritis Impact Measurement Scale (AIMS), and the Symptoms Checklist-90-Revised (SCL-90-R). Pain behavior totals were derived using the Pain Behavior Observation method of McDaniel et al. (1986). Results indicated that the scoring method of Keefe and Block (1982) developed for use with back pain patients was more generalizable to this FS sample. Total pain behavior score using the Keefe and Block scoring method was correlated with self-report pain. Interrater reliability ranged from 0.9 to 0.99. Hierarchical multiple regression analyses revealed that age and physical disability predicted the total pain behaviors. The Pain Behavior Observation method appears a valid pain measurement technique independent of depression among FS patients. Directions for future research are discussed.