Item response tree (IRTree) models are a flexible framework to control self-reported trait measurements for response styles. To this end, IRTree models decompose the responses to rating items into sub-decisions, which are assumed to be made on the basis of either the trait being measured or a response style, whereby the effects of such person…

This study tested five confirmatory factor analytic (CFA) models of the Positive Affect Negative Affect Schedule (PANAS) to provide validity evidence based on its internal structure. A sample of 223 club sport athletes indicated their emotions during the past week. Results revealed that an orthogonal two-factor CFA model, specifying error…

Multisample confirmatory factor analysis (MCFA) and latent mean structures analysis (LMS) were used to test measurement invariance and latent mean differences on the Kaufman Adolescent and Adult Intelligence Scale[TM] (KAIT) across males and females in the standardization sample. MCFA found that the parameters of the KAIT two-factor model were…

An overview of generalizability theory and a FORTRAN computer program for studying the generalizability of scores in a three facet, four factor design are presented. An illustrative example is presented. (Author/JKS)

An efficient algorithm for maximizing split-half reliability coefficients is described. Coefficients derived by the algorithm were found to be generally larger than odd-even split-half coefficients or other internal consistency measures and nearly as large as the largest split half coefficients. MSPLIT, Odd-Even, and Kuder-Richardson-20…

This paper demonstrates that, although aggregated standard scores can correlate substantially with aggregated raw scores, their correlations with an external criterion can differ markedly. This will occur when the variances of the components differ, and these differences are related to the correlations of the components with the criterion. (Author)

Introduces relative mean substitution as an approach to the substitution of missing values in surveys with Likert-type scales. Compares the approach to three other methods for dealing with missing data for samples of 1,674, 400, and 100. Results indicate that the relative mean substitution approach produces the most accurate estimates. (SLD)

The tendency to extreme scores (TES) can affect sensitive indices, such as Cattell's coefficient of pattern similarity, so that a flat profile will, in general, be found more similar to a standard than will an extreme profile. TES is especially critical when profile matching is used in clinical diagnosis. (Author/BW)

GURU is a computer program developed to analyze data generated by open-ended question techniques such as ECHO or other semistructured data collection techniques in which data are categorized. The program provides extensive descriptive statistics and allows extensive flexibility in comparing data. (Author/JKS)

Confusion in the literature between the concepts of internal consistency and homogeneity has led to a misuse of coefficient alpha as an index of item homogeneity. This misuse is discussed and several indices of item homogeneity derived from the model of common factor analysis are offered as alternatives. (Author/JKS)

Focuses on the practical effects of using different statistical treatments with omitted and not-reached items in an item-response theory application. The strategy selected for scoring such items has considerable impact on the interpretation of results for individual or group-level assessments. (Author/SLD)

A computer program is described which provides comprehensive information on test profile differences among independent groups based upon profile centroid separation, shape, elevation, and scatter. Descriptive statistics, univariate and multivariate measures of strength of association, test of homogeneity of covariance matrices, and comparative…

Study based on F. M. Lord's arguments in 1957 and 1959 that tests of the same length do have the same standard error of measurement. (CB)

A general method for identifying the separate components of the Rod-and-Frame Test consists of correlating theoretical patterns of scores with obtained test scores of single subjects. The correlation test calculates probability values from the test data. In this way, fit can be determined between theoretical pattern and test scores. (Author/BW)

The effect of stratified sampling of items based on item difficulty and/or interitem correlations on the estimation of test score distribution parameters using multiple matrix sampling was studied. Results indicated that stratification did not consistently improve the stability of parameter estimation. (Author/JKS)