Based on previous electromyographic studies the 'empty can' (EC) and 'full can' (FC) tests have been promoted as being able to isolate activity to supraspinatus and are therefore diagnostic of supraspinatus dysfunction. However, there is evidence to suggest that these positions do not selectively activate supraspinatus, and that they may have poor diagnostic accuracy, bringing into question the validity of these tests. Due to limitations in the original EMG papers describing the EC and FC tests the current study aimed to further examine shoulder muscle activity during these tests to determine their ability to selectively activate supraspinatus. EMG activity was recorded from 13 muscles in 15 normal subjects while performing these two tests. Results showed that during the EC and FC muscle tests nine and eight other shoulder muscles, respectively, were equally highly activated as supraspinatus. It was concluded from these results that the EC and FC tests do not primarily activate supraspinatus with minimal activation from other shoulder muscles and therefore, do not satisfy basic criteria to be valid diagnostic tools for supraspinatus pathology. Therefore, these tests should not be interpreted as definitive tests for the clinical diagnosis of supraspinatus pathology. They may, however, be beneficial during shoulder muscle strengthening programs.