*2.4. Statistical Analysis*

For all data, descriptive statistics were calculated, and distributions were verified to identify potential outliers. We also checked the distribution to see if assumptions of normality had been violated by a Shapiro–Wilk test. In the case of skewed distribution, differences between the pre- and post-tests were analysed through a Mann–Whitney U as a non-parametric method. For variables that were normally distributed, we applied a repeated measure ANOVA and Bonferroni post hoc. This statistic model was chosen to evaluate the differences within the groups (pre and post phases) and the differences between the groups, specifically the three different equitation disciplines. Differences between endurance, pony games, and show jumping were tested through a Kruskal–Wallis test as a non-parametric method and by the Bonferroni post hoc test of repeated measure ANOVA in the case of normally distributed variables. To assess the differences between horse riders and age-matched reference values, an independent simple *t*-test was utilised. To understand possible relationships between riding experience and pre-test measured variables, Spearman rank correlation or Pearson correlation analyses were applied when proper. Significance was accepted at the level of *p* < 0.05.
