Insights on the Selection of the Coefficient of Variation to Assess Speed Fluctuation in Swimming

Abstract

The aim of this study was to compare swimming speed and speed fluctuations in front crawl between swimmers of different performance levels using discrete variables against statistical parametric mapping (SPM). The sample was composed of 34 male swimmers divided into three groups: (i) group #1—recreational swimmers; (ii) group #2—competitive swimmers aged 12 to 14 years; (iii) group #3—competitive swimmers aged 15 to 17 years. Swimming speed and speed fluctuations (calculated based on four different conditions) were used as discrete variables. Using these discrete variables, ANOVA one-way was used to verify differences between groups, and Bonferroni post-hoc correction for pairwise comparison whenever suitable. SPM (with similar statistical tests) was used to analyze the swimming speed and fluctuation as a continuous variable. Overall, both statistical approaches revealed significant differences (p < 0.001) in swimming speed and speed fluctuations. However, as discrete variables (in four different conditions), the speed fluctuation was not able to detect significant differences between groups #2 and #3. Conversely, SPM was more sensitive and did yield significant differences between these two groups. Therefore, researchers and coaches should be aware that the speed fluctuation as a discrete variable may not identify differences in swimming speed fluctuations when the average value between groups is marginal. On the other hand, SPM was more sensitive in analyzing all groups.