Dear
SPMers,
I am currently
analysing a 2x2 factorial ANOVA at the second level but have run into a problem
of corrected significance using standard main effects and interactions as only
one cell is showing any significant effects. However, if I run an omnibus
F-test looking for changes between the four cells i.e. [1 -1; 0 1 -1;
0 0 1 -1], then I do get corrected significance. I then plan to do 6 post-hoc
tests to find which comparisons are significant. I want to know if this is
ok or will it be frowned upon by paper reviewers?
Thanks
Shane
Shane McKie, Ph.D
University of Manchester