On a sample of 11 volleyball sets played in the Men's Champions League, intrarater and interrater reliability was analyzed in the rating of 503 spikes in attack and counterattack. Three reliability components (rater association, rater bias and rater distribution) were analyzed separately. The reliability of the first rater (the author of the study) was analyzed by employment of the test-retest method. In the process, correlation and differences of the same rater were analyzed at two points of time. This way, the rater's consistency in spike efficacy rating was tested. Repeated measurement was conducted on the same group of sets, four weeks after the first one, in order to avoid any influence of the first measurement exerted on the second measurement. An additional check of the first rater's reliability was conducted by comparison of his rating to the second (expert) rater's rating on the same sample of 11 randomly selected sets at the first point of time. Pearson's correlation coefficient was applied to determine the raters' maximum level of concordance at two points of time, as well as a very high level of concordance between the rater and the expert. The dependent samples t-test disclosed no differences between the raters at two points of time, while the independent samples t-test showed very small differences between the rater and the expert. The very high level of similarity in rating distribution between the raters in repeated measurements and the experts substantiates the very high level of reliability in the rating of events analyzed.
© Copyright 2015 Kinesiologia Slovenica. Faculty of Sport Universität Ljubljana. Alle Rechte vorbehalten.