PG851: GLM--ANOVA

This blog is for the use of students enrolled in either section of PG851. It is intended to be a forum for asking, discussing, clarifying, and helping.

Monday, January 31, 2011

Creating new variables for post hoc?

If there is no significant interaction, do we have to create new variables for the post hoc?

posted by Jennifer @ 7:01 PM  

4 Comments:

  • At 7:55 PM, Anonymous Nikki said…

    No, you would not need to if the interaction effect is not significant. You would still want to do a post hoc for the main effects if they are significant, however.

     

  • At 8:39 PM, Anonymous Anonymous said…

    I’m confused about how to report numbers 5 and 6. I’m showing significant main effects (although not according to Bonferroni for region) but no significant interaction effect.
    Also, I’m not sure about my placement of the descriptive stats for the main effect of gender. Should that be somehow in the table or is it okay to report that in the text since it is only 2?
    - Ashli

     

  • At 8:52 PM, Anonymous Nikki said…

    The APA manual will give you the details of when a table is required.

     

  • At 10:54 PM, Blogger Mari Clements said…

    All the comparisons spas gives you automatically are pairwise ,comparisons. Looking at the two sets of post hocs, think about whether or not pairwise (that is, a vs b, b vs c, and a vs c) captures what seems to be going on with these data or not. If this strategy does not seem to be the best for the data, what does that tell you?

    Also, remember to report the means, the standard deviations, and ns for the model you actually analyzed. The exact model you actually analyzed.

     

Post a Comment

<< Home