To accurately compare electromyographic data from different muscles and different subjects, it is necessary to normalize the integrated data obtained from each muscle. The purpose of this study was to identify the manual muscle testing positions that elicit maximal neural activation (integrated electromyography) of three rotator cuff muscles (supraspinatus, infraspinatus, and subscapularis) and five shoulder synergists (pectoralis major, latissimus dorsi, and anterior, middle, and posterior deltoids). The electromyographic activity of these eight muscles was examined in the nondominant shoulders of nine subjects. Indwelling wire electrodes (supraspinatus, infraspinatus, and subscapularis) and surface adhesive electrodes (pectoralis major, latissimus dorsi, and anterior, middle, and posterior deltoids) were placed. Each subject performed a series of 27 isometric contractions, and optimal tests (maximal neural activation) were identified for each muscle. Four tests were identified that resulted in the maximal neural activation of all eight shoulder muscles: 90 degrees of scapular elevation with -45 degrees of humeral rotation for the supraspinatus, anterior deltoid, and middle deltoid: external rotation at 90 degrees of scapular elevation and -45 degrees of humeral rotation for the infraspinatus and posterior deltoid: internal rotation at 90 degrees of scapular elevation and neutral humeral rotation for the subscapularis and latissimus dorsi: and internal rotation at 0 degree of elevation and neutral rotation for the pectoralis major. These results identify four standard testing positions that will provide reference values for normalization of maximal voluntary contraction for the eight muscles of the shoulder examined in this study. Standardization of these test positions offers normalization guidelines that can be used in future dynamic electromyography studies of the shoulder.