For coaches to qualitatively analyze the performance of sport skills as executed by their students, they must possess an internal image of the desired skill against which to make comparisons (Hoffman, 1983; Pinheiro & Simon, 1992). Leas and Chi (1993) have indicated that there are differences between novice and expert swimming coaches in the internal image of the freestyle stroke. They further reported group differences in their ability to diagnose errors in movement form. Study 1 extended that investigation to include a beginning and advanced skill in downhill snow skiing. Two groups of downhill ski instructors (n=8) certified at Level II (n=4) and Examiner status (n=4) were compared on their knowledge of the prototypical versions of the wedge and open parallel turns. They were subsequently tested on their ability to diagnose errors in incorrectly performed videotape versions of those turns. Two Level H instructors internalized the skill similarly to the level of the Examiners on their knowledge of the prototypical skills, while the other two instructors did not appear to have constructed the same type of prototypical model. On the wedge turn diagnostic task, Level II instructors misdiagnosed 50% of the primary errors in student performance compared to a perfect performance by Examiners. On the open parallel diagnostic task, performance across groups was similar for the primary error. It was suggested that the open parallel level of skiing is similar to the skiing ability of Level II instructors, which may have enhanced their ability to better diagnose the errors associated with that skill compared to the wedge turn. In Study 2, participants prescribed exercises for the errors identified in Study 1. Results indicated that exercises primarily addressed errors in the same order as they were prioritized. The lesson plans of Level II participants, however, attempted to address 35% more errors than Examiners. The use of part-task teaching methods, used by seven of eight participants, was subsequently addressed and theoretical implications were discussed. A possible theory of expertise explaining group differences was described. / Graduation date: 1999