The tests of significance for koppa presented by Kruskal (1958) for one sample and by Warner and Gray (1973) for two samples are reviewed. The two sample test is extended to include multiple comparisons for both a priori and a posteriori analysis problems. A significant difference between two or more koppa values, within different categories of a third variable, demonstrates an interaction effect. Empirical examples are included.
Get full access to this article
View all access options for this article.
References
1.
Frideres, J. S. and Warner, L. G.Social Constraints and Need for Approval as Interactional Variables in Attitude-Action Relationships. Unpublished manuscript.
2.
Glass, G. V. and Stanley, J. C.Statistical methods in education and psychology. Englewood Cliffs: Prentice Hall, 1970.
3.
Hays, W. L.Statistics for psychologists. New York: Holt, Rinehart and Winston, 1963.
4.
Kirk, R. E.Experimental design procedures for the behavioral sciences. Belmont, California: Brooks/Cole Publishing Co., 1968.
5.
Kruskal, W. H.Ordinal measures of association . Journal of the American Statistical Association, 1958, 53, 815-861.
6.
Marascuilo, L. A.Statistical methods for behavioral science research. New York: McGraw-Hill, 1971.
7.
Warner, L. G. and Gray, L. N.A test of significance between two koppas. EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT, Vol. 33, 1973, 817-822.
