Can We Use Global Scoring in a High-Stakes Objective Structured Clinical Examination (OSCE) for Graduating Medical Students?

The aim of the study was to assess the extent to which we could rely on using global scoring in an OSCE for graduating medical students. The School examined 105 graduate-entry final year medical students. There were 10 stations (120 minutes). All stations were developed and reviewed by the discipline head and leaders. All examiners were trained and standardised. Competency- based marking was developed to assess the students’ abilities. Each station assessed 10 competencies, for each of which a score of ‘0’ (failed), ‘1’ (achieved) or ‘2’ (achieved well) was given. A separate independent global scoring system using ‘Fail’, ‘Borderline’, ‘Pass’ and ‘Very Satisfactory’ was also given. Each examiner was instructed to give the global score first before summating the individual competency based scores to minimise bias. The Pearson’s correlation co-efficient between the competency based and global scoring marks; and the reliability score (Cronbach’s alpha) of the OSCE were calculated. The correlation between the competency based and global scoring marks was highly significant (P \u3c 0.001) for all stations (Table 1), ranging between 0.69 and 0.84 (median 0.78). All students (17) who received a ‘Fail’ global score on a station also ‘failed’ in the competency based scoring. The overall reliability score was 0.66. The overall global scoring method can be used alone for a high-stakes OSCE. Potentially this could save valuable time for individual marking, score processing and data entry. However, we would need to consider our defence in the event of an appeal in such a high-stakes examination. Table 1 Pearson’s correlation between the competency based and global scoring marks Station Correlation P value 1 0.84 \u3c 0.001 2 0.78 \u3c 0.001 3 0.68 \u3c 0.001 4 0.82 \u3c 0.001 5 0.69 \u3c 0.001 6 0.83 \u3c 0.001 7 0.71 \u3c 0.001 8 0.83 \u3c 0.001 9 0.73 \u3c 0.001 10 0.79 \u3c 0.00