Removing nonadditivity from two-way tables with one observation per cell
The standard analysis of variance of an r x s stable with one observation per cell, presupposes the assumptions of additive effects, constant variance, and zero interaction. This paper gives a simple graphical procedure, utilizing Tukey's one degree of freedom for nonadditivity, which enables one to find a transformation of the data that will satisfy the above assumptions. An analysis of variance (ANOVA) can then be performed on the transformed scale.
Please refer to publisher version or contact your library.