Finding Statistical Significance with Elite Female Athletes

Abstract

FINDING STATISTICAL SIGNIFICANCE WITH ELITE FEMALE ATHLETES The practical relevance of statistical significance may vary depending on the situation and the population being studied. It is possible that a small change in performance may be meaningful for an elite athlete, but may not reach statistical significance. PURPOSE: To determine the difference in performance between first and fourth place in elite female track and field athletes and determine whether or not these meaningful differences reach statistical significance. METHODS: The percent difference in performance between first and fourth place for the women's track and field events from the last three Olympiads was determined. Average differences were calculated for different event categories (sprints, N = 72; distance, N = 60; throws, N = 48; and jumps, N = 48). An ANOVA with repeated measures (observed power 0.837, partial eta squared 0.183) was administered to determine whether or not these differences reached statistical significance (p ≤ 0.05). RESULTS: The average difference in performance between first and fourth place in women's track and field events for the last three Olympiads was 1.7063% for sprint events, 0.98% for distance events, 5.35% for throws events, and 3.21% for jump events. None of these differences were statistically significant (p > 0.05). CONCLUSIONS: According to these results of this study, it is possible that a decrease in performance as low as 1% is enough to potentially push an elite female track and field athlete from first place to fourth place (gold medal to no medal). Despite the obvious implications of such a difference, it is likely to be overlooked when relying on statistical tools alone. This demonstrates the importance of context when interpreting data. In untrained individuals or non elite athletes, it may be appropriate to label small changes which do not reach statistical significance as not practically relevant or meaningful. However, when measuring change in a high performance setting, small differences which do not approach significance statistically can be very meaningful. PRACTICAL APPLICATIONS: This calls to question the use of .05 criterion alpha to determine a meaningful change in performance in elite athletes and suggests the need for different standards when interpreting data from these types of populations. Additionally, coaches working with elite female athletes who are determining whether or not a specific stimulus will have a meaningful effect on their athletes should consider that small changes in performance could possibly be the difference between winning and losing in a track and field event.