Non-parametric statistics for psychological research.

Size Estimation of Pictures as a Function of Thematic Content for Schizophrenic and Normal Subjects1

“…Hypothesis II was explored further by a direct comparison of the size estimates of the good and poor premorbid groups The Mann-Whitney [/-test (19) was employed for this purpose This nonparametric test is based on the probability that a single score of either group will appear m a given position when all scores of the combined groups are arranged m rank order The null hjrpothesis for size estimates was rejected for all pictures at the following levels of significance Neutral-p < 05, Dominance, Acceptance, Ignoring, Overprotection-p < 01 No differences were found between the two groups in the size estimates of the square A comparison of the variances of the good and poor premorbid groups by the F-ratio revealed a greater variance for the poor premorbid group on the Dominance picture only {p < 02)…”

guided-inquiry based laboratory instruction: investigation of critical thinking skills, problem solving skills, and implementing student roles in chemistry

“…(Hollander and Wolfe, 1973) In parametric methods, the t-test is used for stated significance based on the assumption of normality. However when the sample size is small, an alternative to t-test may be the Mann-Whitney U test which is a better test under certain circumstances even though it reduces the observations to ranks and may be regarded as the test of the randomization type applied to the ranks of the observations (Moses, 1952). Two unrelated samples can be compared using non-parametric the Mann-Whitney U test also called the Wilcoxon rank sum test (Corder and Foreman, 2009 Choi and Hand (2008) scoring matrix.…”

The cognitive implications of role taking behavior1,2

“…Method of analysts For any given variable (eg, Rorschach developmental score) high and low groups of 6's were formed by dividing at the median The RTT scores of these two groups were then compared Because of the nature of the data, it was decided to employ a distribution-free test of significance The Mann-Whitney two-sample test for independent groups (Moses, 1952) is particularly sensitive to differences m location, l e , provides a distributionfree analogue of the t test for differences between means This test of significance was used m all statistical analyses reported in this section The investigator adopted the 05 level of significance at which to reject the hypothesis that the groups were drawn from the same population with regard to scoring on the compared variable…”