Texts often suggest running preliminary tests for homogeneity of variance prior to running an ANOVA. While it has been known for some time that most of the suggested tests are probably not appropriate, they are still being used. This paper is a review of the literature in terms of the implications involved in running preliminary tests in general…

Neglected issues of simultaneous statistical inference and statistical power in survey research applications of the general linear model are reviewed, and it was found that classical hypothesis testing as it is currently applied, is inadequate for the purposes of social research. The intelligent use of statistical inference demands control over…

A statistical method, Fishers Table of t (or t ratio), is used to determine the statistical significance of a comparison of results from pretests and post-tests of knowledge and skill taken by training program participants. A statistically significant improvement can demonstrate training effectiveness. (MS)

Kaufman and Sweet's article on the regression analysis of unbalanced factorial designs (EJ 111 767) is reviewed. A number of errors are noted, and relevant literature is cited. (GDC)

Research hypotheses versus statistical hypotheses, null hypotheses and alternative hypotheses, and statistical significance versus practical significance are described and illustrated. (CP)

A multiple-choice achievement test was constructed in which both cognitive level and degree of abstractness were controlled in test items. By controlling both dimensions, researchers hoped that the variance in achievement scores of earlier research could be accounted for. Results of testing education majors are discussed. (Author/MT)

The use of measures of magnitude of effect has been advocated as a way to go beyond statistical tests of significance and to identify effects of a practical size. They have been used in meta-analysis to combine results of different studies. Describes problems associated with measures of magnitude of effect (particularly study size) and…

A recent study (Blair and Higgins, 1980) indicated a power advantage for the Wilcoxon W Test over student's t-test when calculated from a common mixed-normal sample. Results of the present study indicate that the t-test corrected for outliers shows a superior power curve to the Wilcoxon W.

Occupational theory research has been associated with low statistical power and a high rate of Type II errors. To increase power, a procedure involving the partitioning of the decision region into three sections, based on the decision-theory approach to significance testing, is proposed. (SK)

Compares three methods of analyzing pairwise treatment differences in a multi-treatment experiment via computer simulation techniques. Under the equal n condition, the robustness of the conventional Tukey Wholly Significant Difference test (WSD) to heterogeneous variances was contrasted with two alternate techniques. Under unequal n conditions,…

Two psychological experiments--Anderson and Shanteau (1970), Berkowitz and LePage (1967)--are reanalyzed to present the problem of the relative importance of low Type 1 error probability and high power when answering a research question by testing several statistical hypotheses. (Author/PN)

Methods of assessing effect size and the flows that limit their usefulness are discussed. The various statistical procedures for estimating variance accounted for are based on different statistical models producing sharply differing results. The methods reflect too greatly the particular study characteristics and hence have limited…

Hintzman's 1980 attack on certain analyses in memory research is based on doubtful presumptions, namely, that contingency tables are inherently suspect as evidence for or against scientific conclusions and that pressing this logical argument is in some way an acceptable substitute for empirically examining the conclusions in question. (Author)