A problem that arises frequently in Analysis of Variance is that after rejection of the overall hypothesis, none of the contrasts of interest may be found to be significant when using the available multiple comparison procedures. A procedure for determining the contrast that is responsible for the rejection of the hypothesis is outlined, and the relationship between the "most significant" contrast and the overall test is shown. The implications of this for carrying out multiple comparisons is discussed.
Get full access to this article
View all access options for this article.
References
1.
Draper, N. R. and Smith, H.Applied regression analysis. New York , Wiley, 1966.
2.
Glass, G. V. and Stanley, J. C.Statistical methods in education and psychology, 1970.
3.
Hays, W. L.Statistics. New York: Holt, Rinehart and Winston, 1963.
4.
Morrison, D. F.Multivariate statistical methods . New York: McGraw Hill, 1967.
5.
Myers, J. L.Fundamentals of experimental design. Boston: Allyn and Bacon , 1972.
6.
Rao, C. R.Linear statistical inference and its applications. New York: Wiley, 1965.
7.
Scheffé, H.Analysis of variance. New York: Wiley, 1959 .
