Objectives. To determine whether the Kibler method of scapular position measurement has adequate inter-tester reliability in a muscular population, such as healthy swimmers, in sport-specific positions, and when using side-to-side measures.Design. Same day repeated measures inter-tester reliability study.Setting. Two Western Australian aquatic centers.Participants. Fifteen junior elite swimmers were measured by three investigators.Main outcome measures. The distances between the medial spine of the scapula and T3/4 (Superior Kibler) and the inferior scapula angle and T7/8 (Inferior Kibler) were recorded. Subjects were tested with arms in neutral, hands on hips, abduction, and full elevation.Results. Superior and Inferior Kibler measures had similar reliability, with intra-class correlation coefficients (ICCs) 0.20-0.82 and standard error of measurements (SEMs) 5.3-11.9 mm. Reliability decreased with glenohumeral elevation for Superior Kibler but not for Inferior Kibler. Side-to-side measures were less reliable with ICCs 0.26-0.64 and SEMs 5.8-14.4 mm.Conclusions. Measurement of scapular position in an athletic population appears to be as reliable as in non-athletic populations (SEMs of 4.5-16.5 mm). This level of reliability is barely adequate to confidently detect clinically important differences of 10-15 mm. A sport-specific position such as full flexion may highlight subtle impairments, but has decreased reliability for the Superior Kibler.