Statistical Analysis of the Median Test and the Mann-Whitney U Test

Statistical Analysis of the Median Test and the Mann-Whitney U Test

By Eric U. Oti*, Michael O. Olusola and Perewarebo A. Esemokumo

ABSTRACT
The median test and Mann-Whitney U test are nonparametric methods designed to handle two samples problem, their data are continuous which consist of two mutually independent random samples. They are used to test whether two (or more) independent samples have been drawn from populations with the same median. In this paper, we discussed and analyzed these two methods using the same illustrative example, testing the null hypothesis at 5 percent significance level and it was observed that both the median and Mann-Whitney U test were statistically significant indicating that the two samples of scores earned by students in Statistics department were drawn from populations with equal median scores in the course.

KEYWORDS: Chi-Square, Contingency Table, Population, Rank, Two Samples.

Full Article

Click here to view article in PDF


Language translation

EnglishFrenchGermanItalianPortugueseRussianSpanish

IJAAR IMPACT FACTOR

Call for papers

Research Articles written in English Language are invited from interested researchers in the academic community and other establishments for publication. Authors who wish to submit manuscripts should ensure that the manuscripts have not been submitted elsewhere neither is it under consideration in another journal. The articles should be the original work of the authors. High quality theoretical and empirical original research papers, case studies, review papers, literature reviews, book reviews, conceptual framework, analytical and simulation models, technical note from researchers, academicians, professional, practitioners and students from all over the world are welcomed.

NEWS UPDATE

IJAAR (DOI: 10.46654) is a voting member of CROSSREF; all hosted journals and published articles are assigned with DOI number.

Authors who haven’t submitted their addresses for hard copies collection are advised to do so by visiting:

www.ijaar.org/submit-address