Showing papers on "Bonferroni correction published in 1980"
TL;DR: Five methods of performing pairwise multiple comparisons in repeated measures designs were investigated and Monte Carlo methods revealed that Tukey's WSD leads to an inflated alpha level when the sphericity assumption is not met.
Abstract: Five methods of performing pairwise multiple comparisons in repeated measures designs were investigated. Tukey's Wholly Significant Difference (WSD) test, recommended by most experimental design texts, requires that all differences between pairs of means have a common variance. However, this assumption is equivalent to the sphericity condition that is necessary and sufficient for the validity of the mixed-model approach to the omnibus test. Monte Carlo methods revealed that Tukey's WSD leads to an inflated alpha level when the sphericity assumption is not met. Consideration of both Type I and Type II error rates found in the simulated conditions for the five procedures suggests that a Bonferroni method utilizing a separate error term for each comparison should be employed.
172 citations
TL;DR: In this article, the upper and lower probability bounds of degree two for the union (or intersection) of a sequence of n events are derived using a linear programming algorithm, compared to that suggested by Kounias and Mann and is shown to be the dual of their linear programming formulation.
Abstract: Upper and lower probability bounds of degree two for the union (or intersection) of a sequence of n events are derived using a linear programming algorithm. The approach is compared to that suggested by Kounias and Mann and is shown to be the dual of their linear programming formulation. The new approach is simpler and more efficient.
10 citations
13 Nov 1980
TL;DR: A computer program was written to make the tests between means based on Bonferroni t and also to make multiple comparisons of the standard deviations associated with the means.
Abstract: To ascertain the agreement among laboratories, samples from a single batch of material are analyzed by the different laboratories and results are then compared. A graphical format was designed for presenting the results and for showing which laboratories have significantly different results. The appropriate statistic for simultaneously testing the significance of the differences between several means is Bonferroni t. A computer program was written to make the tests between means based on Bonferroni t and also to make multiple comparisons of the standard deviations associated with the means. The program plots the results and indicates means and standard deviations which are significantly different.