Three problems with stepwise research methods are explored. Computer packages may use incorrect degrees of freedom in stepwise computations. In addition, stepwise methods do not identify correctly the best variable set of a given size. A third problem is that stepwise methods tend to capitalize on sampling error. (SLD)

The use of multiple comparisons in analysis of variance (ANOVA) is discussed. It is argued that experimentwise Type I error rate inflation can be serious and that its influences are often unnoticed in ANOVA applications. Both classical balanced omnibus and orthogonal planned contrast tests inflate experimentwise error to an identifiable maximum.…

In this study the authors investigated the use of 5 (i.e., Claudy, Ezekiel, Olkin-Pratt, Pratt, and Smith) R[squared] correction formulas with the Pearson r[squared]. The authors estimated adjustment bias and precision under 6 x 3 x 6 conditions (i.e., population [rho] values of 0.0, 0.1, 0.3, 0.5, 0.7, and 0.9; population shapes normal, skewness…

Stepwise methods are frequently employed in educational and psychological research, both to select useful subsets of variables and to evaluate the order of importance of variables. Three problems with stepwise applications are explored in some detail. First, computer packages use incorrect degrees of freedom in their stepwise computations,…

The utility of combining confirmatory factor analysis and second-order methods is illustrated in a study of responses of 487 undergraduate and graduate students to the love instrument of C. Hendrick and S. Hendrick. Second-order confirmatory methods allow the researcher to explore complex realities more thoroughly. (SLD)

Most training programs in education and psychology focus on classical test theory techniques for assessing score dependability. This paper discusses generalizability theory and explores its concepts using a small heuristic data set. Generalizability theory subsumes and extends classical test score theory. It is able to estimate the magnitude of…

Effect sizes have been increasingly emphasized in research as more researchers have recognized that: (1) all parametric analyses (t-tests, analyses of variance, etc.) are correlational; (2) effect sizes have played an important role in meta-analytic work; and (3) statistical significance testing is limited in its capacity to inform scientific…

The present paper suggests that multivariate methods ought to be used more frequently in behavioral research and explores the potential consequences of failing to use multivariate methods when these methods are appropriate. The paper explores in detail two reasons why multivariate methods are usually vital. The first is that they limit the…