ABSTRACT
The Mann-Whitney U test is the non-parametric equivalent to the unpaired t test. It is used to compare the medians (or location) of two unmatched samples that are not normally distributed just as the unpaired t test compares the means from normally distributed data. The median value (or middle value of a set of observations ranked in order) is chosen as an alternative to the mean as a measure of position as it is less sensitive to the outlying values in skewed distributions.
