The method of unweighted means (MUM) is an alternative to the least squares method (LSM) for analyzing nonorthogonal data. The MUM was investigated empirically to determine Type I error rates associated with various degrees of imbalance among the cell frequencies. The MUM was found to have acceptable error rates in both univariate and multivariate cases when the ratio of largest to smallest cell frequency, a, is less than or equal to 5. Previously, the MUM was thought to be inappropriate when a - 2. The issue of the power is also addressed. The LSM has more power on average; however, cases for which the MUM has more power are described.
Get full access to this article
View all access options for this article.
References
1.
Betz, M. A. and Elliott, S. D. (1984). The method of unweighted means in multivariate analysis of variance. Journal of Educational Statistics, 9(3), 193-213.
2.
Bock, R. D. (1975). Multivariate statistical methods in behavioral research. Highstown, NJ: McGraw-Hill.
3.
Elliott, S. D. (1981). The harmonic mean method in multivariate analysis of variance. Unpublished doctoral dissertation, Arizona State University.
4.
Horst, P. and Edwards, A. L. (1982). Analysis of nonorthogonal designs: The 2k factorial experiment. Psychological Bulletin, 91, 190-192.
5.
International Mathematical and Statistical Library (IMSL). (1979). Houston: Author.
6.
McKeon, J. J. (1974). F approximations to Hotelling's 12. Biometrika, 61, 381-383.
7.
Overall, J. E. and Spiegel, D. K. (1969). Concerning least squares analysis of experimental data. Psychological Bulletin, 72, 311-322.
8.
Rankin, N. O. (1974). The harmonic mean method for one-way and two-way analyses of variance. Biometrika, 1, 117-122.
9.
Rao, C. R. (1973). Linear statistical inference and its applications (2nd ed.). New York: Wiley.
10.
Snedecor, G. W. and Cochran, W. G. (1980). Statistical methods (7th ed.). Ames, IA: Iowa State University Press.
11.
Timm, N. H. (1975). Multivariate analysis with applications in education and psychology. Belmont, CA: Wadsworth.
12.
Yates, F. (1934). The analysis of multiple classifications with unequal numbers in the different classes. Journal of the American Statistical Association, 29, 51-66.
