A large-sample technique attributable to Leonard Marascuilo, which is based on Henry Scheffe's well-known multiple comparison procedure, was programmed with the SAS System. The focus was on comparisons among independent binomial samples. As in Scheffe's "data snooping" analytic method, the comparisons can be planned or post hoc, they can be simple or complex, and they need not be orthogonal, but none of them will be statistically significant unless the omnibus statistic, which in this case is chi-square rather than F, is statistically significant. This SAS program can be applied when thedesign is not balanced.
Get full access to this article
View all access options for this article.
References
1.
Keppel, G. , Saufley, W. H., & Tokunaga, H. (1992). Introduction to design and analysis: A Student's handbook (2nd ed.). New York: Freeman.
2.
Marascuilo, L. A. (1996). Large-sample multiple comparisons. Psychological Bulletin, 65, 20-290.
3.
Scheffe, H. (1959). The analysis of variance. New York: John Wiley.
4.
Williams, R. H. (1992a). Dunnett-like comparisons among proportions. Educational and Psychological Measurement, 52, 595-597.
5.
Williams, R. H. (1992b). Multiple compaisons among k independent binomial samples. Educational and Psychological Measurement, 52, 103-105.
6.
Williams, R. H. (1992c). Tukey-like pairwise comparisons among proportions. Educational and Psychological Measurement, 52, 913-914.
