Necessary and Sufficient Conditions forFRatios in theL×J×KFactorial Design with Two Repeated Factors

Abstract
Assuming multivariate normality, a necessary and sufficient condition is derived for the validity of each F ratio in an L × J × K factorial design with repeated measurements on the last two factors. The condition is: C'Σ1C = C'Σ2C =…= C'Σ L C = σ2I, where C is the contrast matrix representing the comparison of interest, and Σ i , i = 1, 2, …, L, is a (JK) × (JK) population dispersion matrix.

This publication has 2 references indexed in Scilit: