Randomized controlled trials have proliferated in education, in part because they provide an unbiased estimator for the causal impact of interventions. It is increasingly recognized that many such trials in education have low power to detect an effect if indeed there is one. However, it is less well known that low powered trials tend to…

Multivariate analysis of variance (MANOVA) is a statistical method used to examine group differences on multiple outcomes. This article reports results of a review of MANOVA in gifted education journals between 2011 and 2017 (N = 56). Findings suggest a number of conceptual and procedural misunderstandings about the nature of MANOVA and its…

Missing data are a persistent problem in psychological research. Peer nomination data present a unique missing data problem, because a nominator's nonparticipation results in missing data for other individuals in the study. This study examined the range of effects of systematic nonparticipation on the correlations between peer nomination data when…

Given the American Psychological Association's strong recommendation to always report effect sizes in research, scholars have a responsibility to provide complete information regarding their findings. The purposes of this study were to (a) determine the frequencies with which different effect sizes were reported in published, peer-reviewed…

The present study aimed to identify the scientific research obstacles facing faculty members in the College of Education at Princess Nora bint Abdul Rahman University (PNU) and to determine the differences in the obstacles according to age, academic rank, scientific specialty, marital status, number of completed studies, and time since the last…

Because of the Flynn effect, IQ scores rise as a test norm ages but drop on the introduction of a newly revised test norm. The purpose of the current study was to determine the impact of the Flynn effect on learning disability (LD) diagnoses, the most prevalent special education diagnosis in the United States. Using a longitudinal sample of 875…

It is suggested that the binomial theorem be used to compute the probability that a given number of Type I errors would occur when a group of null hypotheses are true and that this result be used as the level of significance for the test of an overall hypothesis. (Author/BW)

Too often researchers rely upon the classical normal theory parametric tests to analyze non-normal data, even though the tests may not be robust to violations of that assumption. Fligner's class of two-sample tests for scale is an important development because the test is distribution-free and has desirable properties. This paper outlines the…

Communication researchers have recently expressed concern with the lack of statistical power in their literature. Authors propose a method for increasing statistical power: the partitioning of the decision region in three parts. This procedure results in an unambiguous interpretation of nonsignificant results and leads to increased power. (PD)

A dialog between two fictional teachers provides some basic examples of how research that uses approved methodology may provide results that are significant statistically but not significant practically. (PGD)

The second half of a dialogue between two fictional teachers examines the significance of statistical significance in research and considers the factors affecting the extent to which research results provide important or useful information. (PGD)

The normative implications of applying significance tests in evaluation research are examined. The authors conclude that evaluators often make normative decisions, based on the traditional .05 significance level in studies with small samples. Additional reporting of the magnitude of impact, the significance level, and the power of the test is…

A Monte Carlo technique was used to investigate the effect of the differences in covariate means among treatment groups on the significance level and the power of the F-test of the analysis of covariance. (Author/GDC)

States that a method for estimating gender wage gaps by industry yields estimates that vary according to arbitrary choice of omitted reference groups. Suggests alternative methods not susceptible to this problem that can be applied to other contexts, such as racial, union/nonunion, and immigrant/native wage differences. (SK)

Two stage multiple-comparison procedures give an exact solution to problems of power and Type I errors, but require equal sample sizes in the first stage. This paper suggests a method of evaluating the experimentwise Type I error probability when the first stage has unequal sample sizes. (Author/BW)