EVALUATIONS USING NONPARAMETRIC STATISTICS

In this paper the resilience of nonparametric statistics to an extensi on of the sample set is investigated. It is shown that the addition of fourth and fifth samples does influence one's decision about a differ ence in scores between the original samples. This could result in stra nge recommendations like: The experiment pointed out that no significa nt differences were found between the experimental and control group. However, we recommend you to repeat the experiment, because the differ ences over two experimental and two control groups with exactly identi cal results will probably be significant. The cause of the problem men tioned is sought in the ranking of scores and the statistics used. The paper concludes with a number of recommendations.