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.

Sunday, February 20, 2011

Post Hoc

I am a little confused about the post hocs as well... with repeated measures (and all others?) we do not need to run a post hoc if there is no significant interaction, but there is a (or two) main effect(s)?

Thanks!

Brooke

posted by bfullmer @ 3:15 PM  

2 Comments:

  • At 6:02 PM, Anonymous Nikki said…

    It may help to remember the purpose of post hoc comparisons--to determine which groups (levels of a factor) are significantly different from the other groups in that variable.

    So, in any ANOVA analysis, if the independent variable(s) has more than two levels/groups, you will need to run a post hoc to determine which groups are significantly different from each other. This information should also be reported in a write-up so that others know which groups were sign. different from each other.

    If you have an IV with only two levels (boys and girls for example), you would not need a post hoc for this IV b/c you know that if you obtain a significant result, boys are significantly different from girls---there are no other groups to compare in this example.

     

  • At 6:20 PM, Blogger bfullmer said…

    Thanks!

     

Post a Comment

<< Home