Levene’s Test for Homogeneity of Variance
Although the one-way ANOVA F test is relatively robust with respect to the assumption of equal group variances, large departures from this assumption may seriously affect the level of significance and the power of the test. Therefore, various statistical procedures have been developed to test the assumption of homogeneity of variance. Perhaps the most powerful is the modified Levene test [see References 1—4]. You test for the equality of the population variances:
H0: All the population variances are equal; for example, H0: σ1 = σ2 = σ3
against the alternative.
H1: Not all the population variances are equal; for example, σ1 ≠ σ2 or σ1 ≠ σ3 or σ2 ≠ σ3 or σ1 ≠ σ2 ≠ σ3
The Levene test is based on the idea ...
Get Statistics for Six Sigma Green Belts with Minitab and JMP now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
