728
Chapter 16: One-Way Analysis of Variance
Bartlett’s
test and the
Levene’s
test. Bartlett’s test assumes that the data set
is normally distributed and the Levene’s test assumes that the data is
continuous.
Figure 16-12
shows a portion of the output for these tests to
help us determine whether the equal or constant variance assumption has
been violated. Observe that both tests did not reject the null hypothesis for
equal variance for the three data sets.
Figure 16-12:
Bartlett’s and Levene’s Tests for the Three Drugs
Since both the normality and the constant variance assumptions were not
violated, and assuming sample independence one can now proceed to test for
equality of the population means using the analysis of variance procedure.




