In using multiple regression as a data-analysis technique, one problem that might arise is the overuse of a full model with several restricted models, without adjusting the probability level. Such an approach would violate the apparent (or tabled) probability level. This has long been a concern in statistics. The intent of the present paper has been to reconceptualize two of the better known multiple-comparison procedures in a multiple regression approach. The change in the regression approach requires assessing the results of multiple uses of a full model to a correct distribution rather than a straightforward use of the F distribution.
Get full access to this article
View all access options for this article.
References
1.
BottenbergR. A.WardJ. H.Applied multiple linear regression. Lackland Air Force Base, Texas: Personnel Research Laboratory, Aerospace Medical Division, 1963. (PRL-TDR-63-6)
2.
CohenJ.Multiple regression as a general data-analytic system. Psychological Bulletin, 1968, 70, 426–443.
3.
DuncanD. F.Multiple range and multiple F tests. Biometrics, 1955, 11, 1–42.
4.
DunnO. J.Multiple comparisons among means. Journal of the American Statistical Association, 1961, 56, 52–64.
5.
DunnettC. W.A multiple comparison procedure for comparing several treatments with a control. Journal of the American Statistical Association, 1955, 50, 1096–1121.
6.
JenningsE.Fixed effects analysis of variance by regression analysis. Multivariate Behavioral Research, 1967, 2, 95–108.
RyanT. A.The experiment as the unit for computing rates of error. Psychological Bulletin, 1962, 59, 301–305.
12.
SchefféH.A method for judging all contrasts in the analysis of variance. Biometrika, 1953, 40, 87–104.
13.
WardJ. H.Synthesizing regression models—an aid to learning effective problem analysis. American Statistician, 1969, 23, 14–20.
14.
WilliamsJ. D.A multiple regression approach to multiple comparisons for comparing several treatments with a control. Journal of Experimental Education, 1971, 39, 93–96.
15.
WinerB. J.Statistical principles in experimental design. (2nd ed.) New York: McGraw-Hill, 1971.
