Multifactor and multidimensional data quality analysis of judge scoring in diving competition

Area of Science:

• Sports Science
• Statistical Analysis

Background:

• Judge scoring in sports provides performance metrics but faces reliability issues.
• Controversies over judging data quality can impact fairness in competitions.

Purpose of the Study:

• To propose and illustrate a method for evaluating judging data quality in diving events.
• To identify specific areas of inconsistency in judge scoring.

Main Methods:

• Utilized Kendall covariance and Kendall correlation coefficients to analyze judging data.
• Evaluated data quality across overall competition, individual divers, rounds, and diving techniques.
• Examined collective panel quality, inter-judge agreement, and individual judge alignment with final scores.

Main Results:

• The overall data quality of the judge panel's scoring was found to be high.
• Identified specific inconsistencies in scoring for certain divers, rounds, and dives in two case studies.
• Highlighted instances of poor data quality associated with specific judges in Competition 1 and Competition 2.

Conclusions:

• The proposed method effectively uncovers data quality disparities related to judges.
• The Kendall correlation coefficient may have limitations with small sample sizes or minimal data variation.
• The findings underscore the importance of rigorous data quality assessment for ensuring fairness in sports competitions.