Analysis of variance ideally would have been used to identify significant differences in swimming performance and positive psychological development across the three intervention phases (i.e., pre-, post-, 1-month post-intervention). However, not all data sets were complete. Not all participants competed in each swimming stroke, at every distance, and in each experimental timeframe; for example, 10 of the 36 participants competed in two or more 50 m butterfly competitions during the PST program intervention to allow a meaningful comparison between pre- and post-intervention personal best times (see Table 2). Thus, it was necessary to use paired sample t tests to determine the effect of the intervention protocol on improvement in swimming performance and, similarly on each of the positive psychological scales (Heiman, 2002). The squared point-biserial correlation coefficient (r2pb) was calculated to estimate the effect size with each statistical