Skip to Main Content
 

Global Search Box

 
 
 
 

ETD Abstract Container

Abstract Header

An Analysis of the Variation in Dressage Judge Scoring

Abstract Details

2018, Master of Science (MS), Bowling Green State University, Applied Statistics (Math).
In any subjectively scored sport, there is always the possibility of judge bias. After events at the 2008 Olympics at Beijing caused the scoring methods for international dressage competitions to come under scrutiny, the Federation Equestre Internationale (FEI) responded to the need for additional research into the issue. Following the patterns of previous research, their studies relied heavily on techniques such as Analysis of Variance (ANOVA) to make conclusions about contributing factors to judge bias and pointed to factors such as location of the event, the breed of the horse, and the test level as indicators of bias. While ANOVA is helpful for finding variation between groups, it does not take into account the individuality of competitors and different sample sizes. For that reason, in this study we focus on Bayesian multilevel models to examine the variation in dressage judge scoring. Not only do these models allow for individual skill levels to vary between competitors, but they also adjust for different sample sizes when some individuals provide more information than others. In our models, we examined the fixed factors of region, test level, and judge rating, but also allowed varying intercepts for individuals within groups for judge, horse, and rider. Since our focus was judge bias, we used different models to see how outside factors affected variation in judge scores. While adding different factors did show some impact on the variations for the three groups, those effects did not necessarily indicate bias. Instead, using multilevel models implied that most of the variation in dressage scores is due to differences between riders while judges score fairly similarly. Furthermore, the percentage of the overall score that is due to judge variability is quite small compared to the percentage contributed by the horse and rider implying that skill level is the most important factor in dressage scores.
James Albert, Ph.D. (Advisor)
Christopher Rump, PhD. (Committee Member)
John Chen, Ph.D. (Committee Member)
53 p.

Recommended Citations

Citations

  • Kreuz, Kreuz, S. (2018). An Analysis of the Variation in Dressage Judge Scoring [Master's thesis, Bowling Green State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1530480688061936

    APA Style (7th edition)

  • Kreuz, Kreuz, Sarah. An Analysis of the Variation in Dressage Judge Scoring. 2018. Bowling Green State University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1530480688061936.

    MLA Style (8th edition)

  • Kreuz, Kreuz, Sarah. "An Analysis of the Variation in Dressage Judge Scoring." Master's thesis, Bowling Green State University, 2018. http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1530480688061936

    Chicago Manual of Style (17th edition)