Two-Way Analyses of Variance

(NOTE: This outline covers pages 292-310 of Ch. 13. A different outline covers pages 276-292 of this chapter.)

  1. Follow-Up Tests
    1. Follow-up tests to probe significant main effects
      1. Situations in which post hoc tests are and are not needed
      2. Presentation of results

    2. Follow-up tests to probe a significant interaction
      1. What researchers do not do if the interaction is significant
      2. Why main effect means can be misleading in the presence of interaction
      3. Graphing the interaction
      4. Tests of simple main effects
      5. Making all possible pairwise comparisons among the cell means
  2. Planned Comparisons
  3. Assumptions Associated With a Two-Way ANOVA
    1. The four assumptions
    2. Which assumptions can be tested . . . and when they should be tested
    3. Options for the researcher in case one or more assumptions seem untenable
    4. The notion of a "robust" analysis
    5. Biased F-tests
  4. Estimating Effect Size and Conducting Power Analyses in Two-Way ANOVAs
    1. The important distinction between statistical significance and practical significance
    2. Ways to estimate effect size (and criteria for assessing such estimates)
      1. d
      2. eta squared
      3. omega squared
      4. f
    3. A power analysis
      1. Timing
      2. Purpose
  5. The Inflated Type I Error Rate in Factorial ANOVAs
    1. The familywise error rate
    2. The Bonferroni adjustment technique
  6. A Few Warnings Concerning Two-Way ANOVAs
    1. Evaluate the worth of the hypotheses being tested
    2. Remember that two-way ANOVAs focus on means
    3. Remember the possibility of Type I and Type II errors
    4. Be careful when interpreting nonsignificant F-tests



Copyright © 2012

Schuyler W. Huck
All rights reserved.

| Book Info | Author Info |

Site URL: www.readingstats.com

Top | Site Map
Site Design: John W. Taylor V