Test that several groups come from populations with the same variance. For sufficiently large sample sizes, a nonsignificant P value means there is insufficient evidence that the variances differ. The test depends heavily on the observations being from normal distributions.
