OUTLINE FOR CHAPTER
13 (Part 2)
TwoWay Analyses of Variance
(NOTE: This outline covers pages 324345 of Ch. 13.
A different outline covers pages 306324 of this chapter.)
 FollowUp Tests
 Followup tests to probe significant main effects
 Situations in which post hoc tests are and are not needed
 Presentation of results
 Followup tests to probe a significant interaction
 What researchers do not do if the interaction is
significant
 Why main effect means can be misleading in the presence of
interaction
 Graphing the interaction
 Tests of simple main effects
 Making all possible pairwise comparisons among the cell means
 Planned Comparisons
 Assumptions Associated With a TwoWay ANOVA
 The four assumptions
 Which assumptions can be tested . . . and when they should
be tested
 Options for the researcher in case one or more assumptions seem
untenable
 The notion of a "robust" analysis
 Biased Ftests
 Estimating Effect Size and Conducting Power Analyses
in TwoWay ANOVAs
 The important distinction between statistical significance and practical significance
 Ways to estimate effect size (and criteria for assessing such
estimates)
 Measures of association
 Eta squared
 Omega squared
 f
 d
 Power analyses
 Performing a power analysis so as to determine the "proper
sample size"
 Performing a post hoc power analysis
 The Inflated Type I Error Rate in Factorial ANOVAs
 The familywise error rate
 The Bonferroni adjustment technique
 A Few Warnings Concerning TwoWay ANOVAs
 Evaluate the worth of the hypotheses being tested
 Remember that twoway ANOVAs focus on means
 Remember the possibility of Type I and Type II errors
 Be careful when interpreting nonsignificant Ftests
