Two Way ANOVA - SPSS (part 3)

I perform and interpret a two way ANOVA in SPSS. The example consists of 2 between subjects factors: one with 2 levels and one with 3 levels. So, it's a 2 * 3 between subjects design. I test both main effects and the interaction effect. I then follow-up with some basic post-hoc tests. I point out that the proper way to test the simple main effects for the interaction can only be achieved by adding a term to the syntax, as it can not be done through the menus. Also, this example is based on unbalanced design. That is, the sample sizes are unequal. This has important implications for factorial anovas, as I demonstrate through comparisons between the means reported in the descriptives table versus the means reported in the 'estimated marginal means' table.

The references I mention in the video are:


Keselman, H. J. Games, P. A., & Rogan, J. C. (1979). Protecting the overall rate of Type I errors for pairwise comparisons with an omnibus test statistic. Psychological Bulletin, 6(4), 884-888.

Keppel, G. (1982). Design and analysis: A researcher's handbook. New Jersey: Prentice Hall.

Howell, D. C. & McConaughy, S. H. (1982). Nonorthogonal analysis of variance: Putting the question before the answer. Educational and Psychological Measurement, 42, 9-24.

Category:

Education

Tags:

• factorial anova
• 2 way anova
• SPSS
• estimated marginal means
• unequal sample sizes
• unbalanced design