A new method is proposed for a simultaneous factor analysis of dichotomous responses from several groups of individuals. The method makes it possible to compare factor loading pattern, factor variances and covariances, and factor means over groups. Generalized least squares is used as the estimation procedure. (Author/JKS)

The practicality of using the EM algorithm for maximum likelihood estimation of item parameters in the marginal distribution is presented. The EM procedure is shown to apply to general item-response models. (Author/JKS)

Investigated the correspondence between stage changes hypothesized by the Kubler-Ross theory and the perception of the course of illness by seriously ill patients and their spouses. Supported the use of Q-methodology as a research procedure for investigations of terminal illness. (Author)

A multidimensional scaling analysis of 21 work outcomes yielded a latent structure with three dimensions: value attached by society, level of psychological need met, and extent the outcome is inherent in the work itself. Results suggest a multidimensional model of work outcomes is more useful than a categorical approach. (Author)

An empirical investigation of the statistical procedure entitled nonlinear principal components analysis was conducted on a known equation and on measurement data in order to demonstrate the procedure and examine its potential usefulness. This method was suggested by R. Gnanadesikan and based on an early paper of Karl Pearson. (Author/AL)

Teachers' ratings of their students on a 40-item learning style checklist based on M. Rosenberg's descriptive theory of four learning styles were factor analyzed for a predominantly Black sample of 377 eighth grade students. (Author/PHR)

The partial correlation coefficient is derived analytically under exemplary factor patterns. In these patterns, variables are described as an additive composition of a set of orthogonal factors, including general, common, and specific factors. Viewed in this framework, it is evident that the partial correlation may yield spurious results.…

A jackknife-like procedure is developed for producing standard errors of estimate in maximum likelihood factor analysis. Unlike earlier methods based on information theory, the procedure developed is computationally feasible on larger problems. Examples are given to demonstrate the feasibility of the method. (Author/JKS)

The relationship between the Michigan Alcoholism Screening Test (MAST) and the subject's sex and income was significant. If male or poor, he was more likely to receive a higher score on MAST. There was no significant correlation with age, marital status, education, or duration of residency. (Author/BEF)

The dimensional structure of infants' behavioral repertoire was shown to be highly stable over 3 to 15 months of age. Factor analysis of parent questionnaire data produced seven factors named Intensity/Activity, Regularity, Approach-Withdrawal, Sensory Sensitivity, Attentiveness, Manageability and Sensitivity to New Food. An eighth factor,…

A mathematics test and the Thurstone PMA Battery were administered to 225 technical students. The item/item correlations were analyzed using both a principal components and a maximum-likelihood method. After varimax rotation, the same structure emerged from both. Results suggest a "mathematical ability" factor independent of…

Estimators which are optimal under assumptions of normality are shown to be vulnerable to the effects of outliers. A survey of robust alternatives is presented. Included are alternatives to the mean, standard deviation, product-moment correlation, t-test, analysis of variance, multivariate techniques, and schemes for outlier detection. (Author/JKS)

The Activity Vector Analysis is a short personality measure, widely used in industry for personnel selection, placement and assessment. This paper questions the statistical basis of the measure as it has appeared in the literature. Recommendations are made for revision of the measure and for resolution of theoretical incongruencies. (Author/JKS)

The Internal-External Locus of Control Scale was modified to a Likert-type scale and subjected to a principal components analysis and varimax rotation on a sample of subjects. Four factors were extracted which were then related to other personality measures on the same sample. (JKS)

Factor analysis of data from 1,070 federal employees, 575 undergraduates and 575 graduate students, faculty, and staff uncovered some weaknesses in the Merit Systems Protection Board's sexual harassment survey instrument. This type of survey does not adequately measure sexual coercion or quid pro quo forms of harassment. (SK)