Abstract

The adopted selection of test, regardless of what to measure should follow important points. It should be kept in mind that if tests with poor coefficients of validity, reproducibility and objectivity are selected, the evaluation will also be poor and consequently, one won't have acceptable parameters for decision making (Marins & Giannichi, 1996). The aim of this study was to test the validity and reproducibility of the 15 meters test, proposed by Fomitchenko (1996), for Sprint swimmers. Swimmers at state and regional levels were selected in the years 1996, 1997 and 1998, of the teams APAN/ENERGIL-C and CLUBE ESPERIA. The intra-class correlation coefficient was adopted as well as inter-class correlation coefficient with level of significant of p<0,05. It can be verified that the 15 meters test possesses the validity and moderate reproducibility which was high among males. However, among females only a low statistically significant reproducibility was found. Even so, the intra-class correlation coefficients of the 50-meter (51,8%) and 100-meter (40,9%) tests showed good behaviour. The limitation of the female group is in the number of the studied sample (n=7). Therefore, it can be conclued that the 15 meters test fills the pre-requirement of the trustworthy test for sprint swimmers. Verf.-Referat