We used the G*Power package5 to perform a power analysis after data collection. We calculated the power (12 b) of a bilateral comparison by defining the sample size as 19 (for a single supraspinatus tear) or 9 (for multiple [supraspinatus and infraspinatus] tears), the threshold of significance (a) as .05, and the effect size as 0.37 and 0.35 in abduction strength for single and multiple tears, respectively. Also, power of comparison of shoulder pain with and without pathological shoulder conditions was calculated by defining the sample size as 87, a as .05, and effect size as 0.83 in scapular malposition, 0.45 in scapular dyskinesis, and 1.12 on the Hara test.