The problem of testing two correlated proportions with incomplete data is considered by means of Monte Carlo simulations studies. A test proposed in this paper, which can be regarded as a generalization of McNemar's test, is recommended in all cases with incomplete data and not too small samples. (Author)

Examination of the literature on statistics indicates that the rejection of the hypothesis of chance dates from the turn of the century. Early statements about statistical significance were given in terms of the "probable error." These earlier conventions were adopted and restated by Fisher in his "Statistical Methods for Research…

The binomial effect size display (BESD) displays the change in success rate attributable to a treatment procedure. It is readily understandable, applicable in varied contexts, and conveniently computed. (Author/GK)

Analysis of Covariance (ANCOVA) is robust with respect to dual violations of the assumptions of equal regression slopes and normality of distributions provided that group sizes are equal, but displays disruptions of empirical significant levels when unequal regression slopes and unequal group sizes are coupled with nonnormal distributions.…

The dependence of item discrimination index (D) on item difficulty index (p), and the relationship of D and p to the phi coefficient are delineated. A table of critical values of D significant at the .10 and .05 levels, with p ranging from .10 to .90, is presented. (Author/CTM)

A procedure is discussed for testing the significance of the difference in two correlated correlation coefficients, using Fisher's Z-Transformation. The procedure is applicable to a wide range of problems involving tests between dependent correlations and has documented mathematical support when its power curves are examined. (MH)

True effects of the joint dichotomization of two variables are explored, and implications of the correct formulation for generalizations of the results obtained by S. E. Maxwell and H. D. Delaney (1993) are examined. The inflation of apparent effects can occur when only one or two predictor variables is dichotomized. (SLD)

Reviewed statistics textbooks published since 1995 to determine the treatment of effect sizes and statistical significance tests. Each of the 89 textbooks included statistical significance, while only 60 included information on effect sizes. (SLD)

Studied a major problem with null hypothesis significance testing (NHST) and two common misconceptions about NHST and considered how these issues are treated in 12 recent textbooks used in education research methods and statistics classes. Findings show that almost all textbooks do not acknowledge controversies surrounding NHST. (SLD)

Explains that an initial criterion to determine severe discrepancies and relative strengths and weaknesses is significant difference between regressed standard scores. Presents tables as convenient method to determine magnitude of discrepancy required for significance at common reliability and intercorrelation values. Shows prevalences of…

Although stated objectives of multivariate multiple regression and canonical correlation seem different, aspects of the analyses themselves are mathematically equivalent. A numerical example of the similarities is given, with reference to omnibus significance testing, variable weighting schemes, and dimension reduction analysis. (SLD)

D. Trafimow presented an analysis of null hypothesis significance testing (NHST) using Bayes's theorem. Among other points, he concluded that NHST is logically invalid, but that logically valid Bayesian analyses are often not possible. The latter conclusion reflects a fundamental misunderstanding of the nature of Bayesian inference. This view…