An application of the Mann-Whitney 'U' test

Abstract

The present study makes use of the Mann-Whitney 'U' test in order to certify whether the two independent group of units - engineering and non-engineering - have been drawn from the same population of private sector industrial giants in India or from two different populations having the same distribution. The units selected are ranked in terms of their net sales. The man of the ''''statistic comes to 128 and the standard error 26.4. At 10 per cent level of significance, the two limits of the acceptance region are 84.4 and 171.6. The sample 'U' statistic lie within the acceptance region. There is therefore no difference between the population of two group of units - engineering and non-engineering and the distributions are equal. It is possible to extend the analysis when the two groups are not equal to each other.