The current methods available for the computation of the residual interaction in the Latin square (and higher order fractional designs) are rather ad hoc. The algorithms which exist to construct the residual interaction contrasts are complex and confusing. Additionally, it is difficult to generalize them to other types of fractional factorial designs. These contrasts may be developed with the Gram-Schmidt orthonormalization. Extensions of the approach are developed.
Get full access to this article
View all access options for this article.
References
1.
Bock, R. D. (1975). Multivariate statistical methods in behavioral research. New York: McGraw-Hill.
2.
Freund, R. J. and Littell, R. C. (1981). SAS for Linear Models. Cary, NC: SAS Institute.
3.
Kirk, R. E. (1982). Experimental design (2nd Edition). Monterey, CA: Brooks/Cole.
4.
Rao, C. R. (1973). Linear statistical inference and its application (2nd Edition). New York: Wiley.
5.
SAS Institute Inc. (1982). SAS User's Guide: Statistics (1982 Edition). Cary, NC: SAS Institute.
6.
SAS Institute Inc. (1986). SASIIML User's Guide, Version 5 Edition.Cary, NC: SAS Institute.
7.
Timm, N. H. and Carlson, J. E. (1976). Analysis of variance through full rank models. Multivariate Behavioral Research Monographs, 75-1.
8.
Winer, B. J. (1971). Statistical principles in experimental design. New York: McGraw-Hill.
