Analysis of Rater Severity and Evaluation Factors in Oral Assessments for Golf Instructor Certification: Application of the Many-Facet Rasch Measurement Model

This study applied the Many-Facet Rasch Measurement Model to analyze both the severity levels of raters and the characteristics of evaluation factors in the oral assessment for golf instructor certification. Based on the results, the following conclusions were reached. First, analysis of the vertical facet map showed that Rater D was the most severe and Rater A the most lenient. Candidate 15 had the highest ability, while Candidates 18, 20, and 28 had the lowest. Among evaluation factors, accuracy was more difficult than attitude and clarity of explanation. The rating scale ranged from 1 to 5 points. Second, rater severity and factor difficulty analysis confirmed that Rater D scored most strictly, while Rater A remained the most lenient. Raters C, B, and A demonstrated consistent scoring, but Rater D exceeded acceptable infit and outfit thresholds, indicating inconsistency. Accuracy was statistically more difficult than attitude and clarity, with a significant difference. Third, Candidate 15 showed the highest ability index, followed by Candidates 2, 22, 33, and others with equal scores. Six candidates (e.g., 24, 13) exceeded 1.50 in both infit and outfit values, indicating low response consistency. Overall, the oral assessment effectively distinguished between ability levels. Fourth, only Rater D showed statistically significant bias by assigning higher scores for attitude and clarity and lower scores for accuracy. Other raters did not demonstrate significant bias.