Your data analysis solution
Name is required.
Email address is required.
Invalid email address
Answer is required.
Exceeding max length of 5KB

Identical F value for main effect and interaction

By James Kleiss | Jun 23, 2016 11:47PM CEST

I have an unbalanced mixed design with 23 subjects in one group and 59 in another. Each group is measured in a pre- and post-test situation. I used the Repeated Measures ANOVA with Group, Time and Group*Time as fixed effects, Time as the repeated effect and Subjects as the subject factor. There are two puzzling things about the results: First, the F value is identical for the Group main effect and the Group*Time interaction. I can't get me head around how that can be.

Second, there is main effect of Time even though there is no difference in Time for the smaller group. Time1 for the larger group is larger than either Time for the smaller group and Time2 for the larger group is larger than that. Apparently, the larger value for Time2 with the larger group is sufficient to drive a main effect of Time1. My question is, why isn't there also an interaction? I did check variances and the variance for Time2 with the smaller group is notably larger than variances for the other three cells in the model. Is that enough to mess up the interaction of Time and Group while maintaining a main effect of Time?

By the way, I'm using Type III SS. Can someone help me understand this?

Thanks!

Jim

Up

0

Down

By Thomas | Jun 24, 2016 01:42PM CEST | XLSTAT Agent

Dear James,

Could you send your excel file at support@xlstat.com so that we can give you an answer?

Regards,

This question has received the maximum number of answers.

Contact Us

1c26995d494fb3061dd0ae8571ffc0a4@xlstat.desk-mail.com
https://cdn.desk.com/
false
desk
Loading
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
about
false
Invalid characters found
/customer/portal/articles/autocomplete
9283