Examine the interrater reliability of cervicothoracic and shoulder physical examination in patients with a primary complaint of shoulder pain.Single-group repeated-measures design for interrater reliability.Orthopaedic physical therapy clinics.Twenty-one patients with a primary complaint of shoulder pain underwent a standardized examination by a physical therapist (PT). A PT conducted the first examination and one of two additional PTs conducted the 2nd examination.The Cohen κ and weighted κ were used to calculate the interrater reliability of ordinal level data. Intraclass correlation coefficients model 2,1 (ICC2,1) and the 95% confidence intervals were calculated to determine the interrater reliability.The kappa coefficients ranged from −.24 to .83 for the mobility assessment of the glenohumeral, acromioclavicular and sternoclavicular joints. The kappa coefficients ranged from −.20 to .58 for joint mobility assessment of the cervical and thoracic spine. The kappa coefficients ranged from .23 to 1.0 for special tests of the shoulder and cervical spine.The present study reported the reliability of a comprehensive upper quarter physical examination for a group of patients with a primary report of shoulder pain. The reliability varied considerably for the cervical and shoulder examination and was significantly higher for the examination of muscle length and cervical range of motion.