Biased analyses in relative age effects: The case of Turkish youth archers


Relative age effects (RAEs) occur in sport when one's birthdate leads to participation and/or performance advantages. Some researchers indicate there are biases in RAE analyses, where athletes' birthdate distributions are compared to the general population. Our purpose was to assess RAEs using two types of analyses. Birthdates of 5,034 Turkish youth archers were analyzed. Birthdates were grouped into four quartiles based on the Turkish Archery Federation's selection year: Q1 = January-March…Q4 = October-December. Our analyses focused on three age divisions: 9- to 10-year olds; 11- to 12-year-olds; 13- to 14-year-olds. For each age division, we compared the birthdate distributions of each quartile to the live birth rates in the Turkish population. Following this, we assessed RAEs in the two older age divisions by comparing the birthdate distributions in each quartile to the birthdate distributions of the age group immediately preceding it (e.g., comparing 13- to 14-year-olds to 11- to 12-year-olds). When implementing a traditional RAE analysis (i.e., comparing to live births), all age divisions demonstrated significant RAEs with overrepresentations of relatively older athletes. When using the non-traditional analysis (i.e., comparing to previous age division), there were no significant RAEs in the 13- to 14-year-olds, and a significant RAE favoring relatively younger athletes among 11- to 12-year-olds. Researchers should carefully consider their analyses, as traditional methods might lead to false conclusions about RAEs. Meanwhile, organizations ought to evaluate why relatively younger athletes do not enroll in sport at the earliest entry ages, as this enrollment bias appears to contribute to RAEs.