In 1881, Donald MacAlister posed a problem in the "Educational Times" that remains relevant today. The problem centers on the statistical evidence for the effectiveness of a treatment based on a comparison between two proportions. A brief historical sketch is followed by a discussion of two default Bayesian solutions, one based on a…

This article presents a method to evaluate pretest effects on posttest scores in the absence of an un-pretested control group using published results of pretesting effects due to Willson and Putnam. Confidence intervals around the expected theoretical gain due to pretesting are computed, and observed gains or differential gains are compared with…

Classification procedures are common and useful in behavioral, educational, social, and managerial research. Supervised classification techniques such as discriminant function analysis assume training data are perfectly classified when estimating parameters or classifying. In contrast, unsupervised classification techniques such as finite mixture…

Suggests two complementary frameworks for providing validity evidence for rating data. One conceptualizes raters as data collection instruments that should be subject to traditional procedures for establishing validity evidence; the other evaluates studies using raters from an experimental design perspective allowing the assessment of the study's…

Effects of choice of data analysis program on outcomes of meta-analysis were studied by considering the extent to which four commonly used Schmidt and Hunter validity generalization-based meta-analytic software programs produce identical outcomes. Differences were minor and did not influence outcomes. (SLD)

This study describes how much and in what ways authors of research studies fail to include adequate information about data collection. Results based on analysis of 220 articles from 22 journals show that the quality of measurement reporting continues to be a problem. (SLD)