An empirical study of the cube - root test for homogeneity of variance with respect to the effects of non - normality and power

Monte Carlo techniques were employed to assess the relative effects of non-normality upon an exact test, two Bartlett tests, and Samiuddin's cube-root test for homogeneity of variance. Inaddition, all four tests were compared with respect to empirical power. Though all four tests were grossly affected by departures from normality, the exact test appeared to be the least disturbed. With respect to empirical power, for k = 2, the exact test displayed the greatest power. The Bartlett tests performed comparably to one another, and both were consistently superior to the cube-root test. For k=4 and k=6, the Bartlett tests continued to display comparable power, and both were consistently superior to Samiuddin's test.