Compared approaches to determining the precision of gain scores: (1) the asymptotic normal distribution of the maximum likelihood estimator of the person parameter; and (2) the exact conditional distribution of the gain score. Use of three data sets illustrates that these methods yield more relevant and more detailed information than traditional…

This paper makes the point that descriptive and inferential measures of nonnormality and graphic displays of the frequency distribution of important variables should be routine in research reporting. This point is particularly true for research involving measures with nonarbitrary metrics where the distribution shape is unaffected by measurement…

This study, using data for 200 psychiatric/chemical dependency patients, attempted to justify subscales of the Minnesota Multiphasic Personality Inventory (MMPI). Distributions of all possible split-half correlations for certain research scales of the MMPI revealed negative skewness resulting in spuriously lowered reliability estimates. The scales…

Establishes a type-token identity relating type-token ratio and bilogarithmic type-token ratio; uses the Simon-Yule model of Zipf's law to drive the type-token ratio and provide a statistical model of text generation; and refines the model to allow for decreasing entry rates of new words. (26 references) (Author/CLB)

A generalization of the usual random-effects model based on trimmed means is proposed. The resulting test of no differences among J randomly sampled groups has advantages in terms of Type I errors and can yield gains in power when distributions have heavy tails and outliers. (SLD)

Discusses the shape of the obsolescence function based on the age data for cited journals. An analysis of the aging function is discussed; and statistical fits and mathematical explanations are illustrated, including the model of Avramescu, the Negative Binomial Distribution, the Weibull distribution, the lognormal distribution, and the utility…

This study demonstrates that the number of references at the end of scientific research papers forms a strongly unimodal frequency distribution and that the generalized Gaussian-Poisson distribution fits such unimodal data extremely well. (three references) (Author/MES)

Examination of the stochastics of moderated multiple regression (MMR) reveals that MMR is an appropriate technique when predictors are fixed variables and the distribution of errors is normal but is not appropriate when predictors are random variables and the joint distribution of criterion and predictor variables is multivariate normal. (SLD)

Discusses the use of the generalized inverse Gaussian-Poisson (GIGP) distribution in bibliometric studies. The main types of size-frequency distributions are described, bibliometric distributions in logarithms are examined; parameter estimation is discussed; and goodness-of-fit tests are considered. Examples of applications are included. (17…

Investigates the Leimkuhler curve and the Gini index from a probabilistic viewpoint for studying the concentration of bibliometric distributions. The importance of the time parameter and the special nature of the "nonproducers" in bibliometric studies are examined, standard models of bibliometric processes are described, and further…

A person-fit method that allows researchers to identify nonfitting response vectors is the l(z) statistic. Simulation results show that l(z) may not perform as expected when estimated person parameters are used rather than true person parameters. Other considerations in using true and estimated person parameters are discussed. (SLD)

Uses a confirmatory factor analytic (CFA) model as a paradigmatic basis for the comparison of three widely used structural equation modeling computer programs: (1) AMOS 4.0; (2) EQS 6; and (3) LISREL 8. Comparisons focus on aspects of programs that bear on the specification and testing of CFA models and the treatment of incomplete, nonnormally…

Studied the effects of sample size on the cumulative distribution of selected fit indices using Monte Carlo simulation. Generally, the comparative fit index exhibited very stable patterns and was less influenced by sample size or data types than were other fit indices. (SLD)

Describes a Windows program for checking the suitability of unidimensional logistic item response models for binary and ordered polytomous responses with respect to a given set of data. The program is based on predicting the observed test score distributions from the item characteristic curves. (SLD)

Illustrates a number of ways in which confidence intervals for reliability coefficients can be estimated. Suggests that authors who submit articles to "Educational and Psychological Measurement" report confidence intervals for reliability estimates whenever they report score reliabilities and that they note the interval estimation…