I have a 5(A)x 3(B) x 2(C) design. When dealing with A and B only, there is a main effect of A and B but not of the interaction of A and B. However adding C in gives a main effect of the interaction of A and B. Specifically, everything is significant, except for the interaction of A, B and C. What does the significance of the interaction of A and B in the full model tell me? Is this more accurate than in the A and B only model where the interaction was insignificant?

  • $\begingroup$ Are there equal Ns in every cell of the design? $\endgroup$
    – John
    Dec 8, 2015 at 2:52


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.