See TM 501 201-2 for related articles. (MLP)

Comment on EJ 060 883, and J. Cohen, Statistical power analysis for the behavioral sciences,'' New York: Academic Press, 1969. (CB)

The difference in statistical power between the original Bonferroni and five modified Bonferroni procedures that control the overall Type I error rate is examined in the context of a correlation matrix where multiple null hypotheses are tested. Power differences of less than 0.05 were typically observed for the modified Bonferroni procedures. (SLD)

Results from classical power analysis (Brewer, 1972) suggest that a researcher should not set a=p (when p is less than a) in a posteriori fashion when a study yields statistically significant results because of a resulting decrease in power. The purpose of the present report is to use Bayesian theory in examining the validity of this…

See TM 501 201-3 and EJ 060 883 for related articles. (MLP)

Power and stability of Type I error rates are investigated for the Bartlett and Kendall test of homogeneity of variance with varying subsample sizes under conditions of normality and nonnormality. The test is shown to be robust to violations of the assumption of normality when sampling is from a leptokurtic population. Suggestions for selecting…

The Integrated Moving Average (IMA) model of time series, and the analysis of intervention effects based on it, assume random shocks which are normally distributed. To determine the robustness of the analysis to violations of this assumption, empirical sampling methods were employed. Samples were generated from three populations; normal,…

A critique of EJ 060 883. (CB)

Reluctance to use group mean as the unit of analysis is partly due to the belief that fewer observations per treatment greatly reduces the probability of detecting a treatment effect. This is discussed; equations are presented to facilitate power estimates when the group mean is the unit of analysis. (Author/BW)

The problem addressed is of assessing the loss of power which results from keeping the probability that at least one Type I error will occur in a family of N statistical tests at a tolerably low level. (Author/BW)

A concise table is presented based on a general measure of magnitude of effect which allows direct determinations of statistical power over a practical range of values and alpha levels. The table also facilitates the setting of the research sample size needed to provide a given degree of power. (Author/CM)

A Monte Carlo simulation examined the Type I error rates and power of four tests of the null hypothesis that a correlation matrix equals the identity matrix. The procedure of C. J. Brien and others (1984) was found to be the most powerful test maintaining stable empirical alpha values. (SLD)

Two bootstrap tests are proposed to test the independence hypothesis in a two-way cross table. Monte Carlo studies are used to compare the traditional asymptotic test with these bootstrap methods, and the bootstrap methods are found superior in two ways: control of Type I error and statistical power. (SLD)

The robustness with respect to Type I error and the power of a proposed test statistic in testing the conjoint hypotheses of mean and variability equality were examined in this simulation study. The conjoint test utilizes the maximum p-value from separate tests of equality of means and equality of variability as its p-value to control the Type I…

Contrasts or comparisons can be used to investigate specific differences between means. Contrasts, as explained by B. Thompson (1985, 1994) are coding vectors that mathematically express hypotheses. The most basic categories of contrasts are planned and unplanned. The purpose of this paper is to explain the relative advantages of using planned…