This article first outlines the logic of null hypothesis testing and the problems of using it to evaluate special education research. It then presents three alternative metrics, a binomial effect size display, a relative risk ratio, and an odds ratio, that can better identify important treatment effects using illustrative data from recently…

Data were analyzed to see if a correlation or relationship existed between students' grades in a course and students' rankings of the professor's teaching. Graphing calculators allowed students to visualize relationships between the data and to compute the equation of the regression line, compute the correlation coefficient, and test statistical…

Effect sizes (N=263) were calculated for 165 educational programs reviewed by the Department of Education's Joint Dissemination Review Panel during 1980-83. The overall mean effect size was 0.89. Descriptive statistics and regression analyses determined that effect sizes were related to features of the educational programs and their evaluations.…

This article compares three models that have been the foundation for approaches to the analysis of statistical significance in early childhood research--the Fisherian and the Neyman-Pearson models (both considered "classical" approaches), and the Bayesian model. The article concludes that all three models have a place in the analysis of research…

This article explains and illustrates the estimation of the power of statistical tests used to analyze data in early childhood special education research, and discusses advantages and disadvantages of various ways to increase power, such as using a directional alternate hypothesis or using a parametric, rather than nonparametric, statistical test.…

This tutorial summarizes some of the widely known limitations of tests of statistical significance and then focuses on extracting measures of variation accounted for as a supplement to significance testing. Two measures of variation accounted for, eta squared and omega squared, are discussed. Computational formulas, computational examples, and…

The use of p-values in combining results of studies often involves studies that are potentially aberrant. This paper proposes a combined test that permits trimming some of the extreme p-values. The trimmed statistic is based on an inverse cumulative normal transformation of the ordered p-values. (SLD)

Statistical significance does not necessarily make research meaningful. The problems and implications must be significant and valuable, and the research process should not be emphasized more than product. Qualitative methods and better reporting of conclusions would enhance significance. (SK)

The disparity between methodology and empirical research literature is examined to suggest a cause for the casual acceptance of null hypotheses simply because obtained differences are not statistically significant. Analysis of training for psychological researchers suggests that programs have not emphasized the comprehensive evaluation of all…

The test proposed by Brien, Venables, and Mayo (1984), endorsed by Silver and Dunlap, and supported by computer software that they developed is not a proper test for multivariate independence. A revised appraisal is suggested for the Silver and Dunlap results. (SLD)

Researchers, journal editors, textbook authors, and those responsible for writing publication manuals must work together to enhance the thoughtful reporting of statistical results and to make clear the necessity for reporting effect sizes. (SLD)

Suggests that researchers should report the results of appropriate significance tests and the effect sizes associated with each test. Discusses the roles of textbook authors, publication manuals, and journal editors in leading the movement to better statistical reporting. (SLD)

Documents the recent use of one-tailed F-tests in communication journals, and examines the arguments both for and against their use. Examines the use of these tests within the broader unresolved controversy surrounding the use of one-tailed tests. Recommends that future researchers should most often avoid one-tailed Fs, and generally exercise…

Shares three ideas concerning how to evaluate the practical importance of developmental findings to make them more useful to policymakers: (1) statistical significance tests need to be accompanied by effect size estimates; (2) meta-analyses are helping in using all existing data when examining issues involved in policy debates; and (3) researchers…

During the last several years, researchers have been garnering evidence concerning the benefits of service-learning, and urging its inclusion in the curricular mainstream. This study presents evidence that college students manifest long-lasting, if modest, academic benefits from participating in service-learning. The academic records of 477…