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.

Tuesday, February 05, 2008

equal cell size

Hi,

Can i treat the cell size as equal if they are not over 4:1? This related to the use of what types of sums of squares and post hoc comparison.

Thanks
Esther

posted by Unknown @ 1:44 PM  

1 Comments:

  • At 8:54 AM, Blogger Mari said…

    The choice of types of sums of squares depends upon why the cells aren't equal. If you designed the study to have equal cell size, and the deviations for absolutely equal are due to random dropout, Type III SS are still completely appropriate.

     

Post a Comment

<< Home