Literaturnachweis - Detailanzeige
Autor/inn/en | Supattathum, Suchada; und weitere |
---|---|
Titel | Statistical Power of Modified Bonferroni Methods. |
Quelle | (1994), (24 Seiten)
PDF als Volltext |
Sprache | englisch |
Dokumenttyp | gedruckt; online; Monographie |
Schlagwörter | Correlation; Hypothesis Testing; Matrices; Power (Statistics); Research Methodology; Simulation; Statistical Analysis; Statistical Studies |
Abstract | Multiple-hypothesis testing in the context of a correlation matrix is used to compare the statistical power of the original Bonferroni with six modified Bonferroni procedures that control the overall Type I error rate. Three definitions of statistical power are considered: (1) the ability to detect at least one true relationship; (2) the ability to detect all true relationships; and (3) the average power to detect true relationships. Simulation results show no difference between the seven methods in detecting at least one true relationship, but all six modified Bonferroni procedures are more powerful than the original Bonferroni procedure for the detection of all-true-relationship power and average power. Among the six modified Bonferroni procedures, small differences were observed, with the Holm procedure having the lower power, and the Rom and the Holland-Copenhaver (step-up methods having the highest power. Included are 9 tables and 11 figures. (Contains 11 references.) (Author/SLD) |
Erfasst von | ERIC (Education Resources Information Center), Washington, DC |