Assessing the tendency of judging bias in student competition: a data mining approach

Abstract

PurposeJudging bias is ironically an inherent risk in every competition, which might threaten the fairness and legitimacy of the competition. The patriotism effect represents one source of judging bias as the judge favors contestants who share the same sentiments, such as the nationalistic, racial, or cultural aspects. This study attempts to explore this type of judging bias in a university student competition. In addition, this study tries to expand the literature on judging bias by proposing the term universitarian bias as the judge is suspected to give a higher score to contestants from the same university.Design/methodology/approachThe association rule of data mining is used to accomplish the objective of the study. To demonstrate the applicability of the method, the data set from the annual national university student competition in Indonesia is exploited.FindingsThe result strongly discovers that the universitarian bias is likely to be present. Some recommendations are also provided in order to minimize the bias that might happen again in the future.Practical implicationsAs the implication of the presence of the universitarian bias, the committee should remove all the university features attributed to the participants. This endeavor is expected to minimize the universitarian bias that might happen.Originality/valueThis research is claimed to be the first attempt in implementing the data mining technique in the field of judging bias.