Testing+the+hypothesis

+ If the null hypothesis holds, then there should not be much variation between the categories relative to the variation within the categories

+ The two estimates of variation (between and within) should be roughly equal

+ The larger the difference between the two estimates, the more likely we will be to reject the null hypothesis

+ If category means are about the same value, the differences will not be significant. But the larger the differences between category means and the more homogeneous the categories, the more likely that the differences will be statistically significant