ABSTRACT
PURPOSE: This study aims to abstract the differences of scores between the clinical instructors and standardized patients (SPs) in a clinical performance examination (CPX) using SPs and to correlate the concordance between the evaluation scores and the school records. METHODS: The CPX was administered in 2003 to a total of thirty-six fifth year medical students at Gachon Medical School. The examination consisted of four cases, and four stations were duplicated, each requiring a total of 7.5 minutes per station. Evaluation of the student' s performances was conducted by both clinical instructors and SPs using a formatted checklist. Results were analysed by t-test, agreement rates, and Pearson correlation. RESULT: The mean scores given out by the clinical instructors and the SPs for the newly developed case were significantly different, while those scores for pre-existing cases were not in the old cases. The correlation coefficients between these two evaluation groups were relatively high. And agreements between the two evaluation groups were 0.37~0.72. The mean scores among clinical instructors were not significantly different, but the correlation coefficients and agreement rates were relatively high. The correlation between the evaluation scores and school records did not correlate significantly. CONCLUSION: It is concluded that the CPX is a useful tool to measure the students' essential competences in areas of knowledge, skills and attitude during the subinternship stage. In conducting a successful CPX, it is crucial to reconsider the recycling of cases and the selection and training of SPs aside from the development of an objective checklist.