The new emphasis on performance-based assessment holds much promise for assessment in physical education, particularly for the assessment of motor skills. Physical educators want to know if students can perform motor skills and use those motor skills in game and activity settings. However, the viability of performance-based assessment as a formal measure of complex motor skill performance in physical education has not been investigated. It is not clear whether teacher judgments are credible sources of reliable information for more formal and standardized testing. The purpose of this study was to investigate the accuracy of teacher motor performance data submitted by teachers to the South Carolina Physical Education Assessment Program (SCPEAP). It was also the purpose of this study to describe teacher compliance with the testing protocols of the state assessment program, factors related to teachers' ability to use scoring rubrics accurately, and factors related to data compliance. Specific factors related to teacher accuracy and compliance included: (1) teacher gender; (2) training sessions attended; (3) Physical Education Institute (PEI) in-service sessions attended; (4) the activity assessed; and (5) the final student competency score given to a class. Specific research questions guiding this study were: (1) Can teachers use scoring rubrics to accurately assess motor performance?; (2) What factors affect teachers' ability to use scoring rubrics accurately?; (3) What were the differences between teacher scores and monitoring committee scores?; and (4) What factors were related to data compliance? (Contains 4 tables.)