Question: One assumption for parametric tests with independent measures data is that different treatment conditions have the same variance (homogeneity of variance assumption). However, a treatment effect that increases the mean often also increases the variability. In this situation, the parametric t-test or ANOVA is not justified and Mann Whitney should be used. The following data represent and example of this situation:

A) Compute the mean and variance for each sample. Note the difference between the two sample variances

B) Use the Mann Whitney test with an alpha=0.05 to test for significant differences between the two treatments.