An asymptotic expression for the reliability of a linearly equated test is developed using normal theory. Reliability is expressed as the product of test reliability before equating and an adjustment term that is a function of the sample sizes used to estimate the linear equating transformation. The approach is illustrated. (SLD)

Computer simulations indicate that the correlation between baseline and change, by itself, does not invalidate the use of gain scores to measure change, but when the negative correlation is accompanied by decrease in variance from pretest to posttest, covariance is a superior measure of change. (SLD)

Alternative tests are presented for the between-by-within interaction null hypothesis and for two within-subjects main effects null hypothesis in a split plot design. Estimated Type I error rates for the interaction tests and for several tests of the second null hypothesis are reported. (SLD)

Focuses on regression effects that arise from measurement error, and examines methods used to evaluate pre- to post-therapy score changes that attempt to take these regression effects into account. Discusses statistical and practical problems, identifies and analyzes specific examples, and distinguishes objectives of these methods from other…

A Monte Carlo study was conducted to investigate the robustness of the assumed error distribution in maximum likelihood estimation models for multidimensional scaling. Results show that violations of the assumed error distribution have virtually no effect on the estimated distance parameters. (SLD)

Schouten and Kirkpatrick's review of evaluations of the Miller Assessment for Preschoolers' predictive validity reveals unacceptably high error rates in target population identification. Miller's response suggests flaws in their argument, Humphry and King-Thomas suggest alternative interpretations, and Schouten and Kirkpatrick respond to their…

TESTAT is a supplementary module for the popular SYSTAT statistical package for the personal computer. The program performs test analyses based on classical test theory and item response theory. Limitations and advantages are discussed. (SLD)

Ninety-five subjects observing videotaped secretaries performing well or poorly were told either to evaluate or monitor job performance. Ratings were influenced by the rater's objective. Regardless of objective, results showed a high level of covariance among the independent traits of the secretaries. (SK)

The standard error (SE) of correlations adjusted for selection with commonly used formulas was investigated. The study provides large-sample approximations of SE using the Pearson-Lawley three-variable correction formula, examines the SE under specific conditions, and compares various estimates of SEs under direct and indirect selection. (TJH)

Explores the ways in which error in measurement related to educational standards and the classification of people in educational settings is obscured in most of the practical events involving the assessment of individuals. Establishes the centrality of the measurement of educational standards to the "production" and control of the individual in…

Expands upon the concepts of precision and accuracy at a level suitable for general chemistry. Serves as a bridge to the more extensive treatments in analytical chemistry textbooks and the advanced literature on error analysis. Contains 22 references. (DDR)

Shares examples of reports of uncertainties that exceed the guideline of one or two significant figures. Points out that scientists should take great care in deciding on the appropriate form for reporting their results. (DDR)

Describes procedures for calculating conditional standard error of measurement (CSEM) and reliability of scale scores and classification of consistency of performance levels. Applied these procedures to data from the American College Testing Program's Work Keys Writing Assessment with sample sizes of 7,097, 1,035, and 1,793. Results show that the…

This simulation study investigated the robustness of structural equation modeling to different degrees of nonnormality under 2 estimation methods, generalized least squares and maximum likelihood, and 4 sample sizes, 100, 250, 500, and 1,000. Each of the slight and severe nonnormality degrees was comprised of pure skewness, pure kurtosis, and both…

Missing data are a common practical problem for longitudinal designs. Time-series analysis is a longitudinal method that involves a large number of observations on a single unit. Four different missing-data methods (deletion, mean substitution, mean of adjacent observations, and maximum likelihood estimation) were evaluated. Computer-generated…