Skip to content

MANE 6313

Week 4, Module D

Student Learning Outcome

Analyze simple comparative experiments and experiments with a single factor.

Module Learning Outcome

Assess the assumption of equal variances in one-way analysis of variance.


Test for Equality of Variances

  • Bartlett's Test

  • Relies on the assumption of Normality

  • Levene's Test

  • More robust to departures of Normality

  • R performs both tests!


Bartlett's Test Example Problem

Module D Data

The source for problems 12. 10 and 12.30 is Ostle, Turner, Hicks and McElrath1


Box Plots

Box Plot of Data


Analysis of Variance

Analysis of Variance for 12.30


Residual Plot

Residual Plot


Bartlett's Test

\[ \begin{eqnarray*} H_0&:&\sigma^2_1=\sigma^2_2=\cdots=\sigma^2_a\\ H_1&:&\mbox{above not true for at least one }\sigma^2_i \end{eqnarray*} \]

Bartlett's Test


Levene's Test

\[ \begin{eqnarray*} H_0&:\sigma^2_1=\sigma^2_2=\cdots=\sigma^2_a\\ H_1&:\mbox{above not true for at least one }\sigma^2_i \end{eqnarray*} \]

Levene's Test


R Demonstration



  1. Ostle, Turner, Hicks and McElrath. (1996). Engineering Statistics: The Industrial Experience. Duxbury Press.