A mastery test is frequently described as follows: an examinee responds to n dichotomously scored test items. Depending upon the examinee's observed (number correct) score, a mastery decision is made and the examinee is advanced to the next level of instruction. Otherwise, a nonmastery decision is made and the examinee is given remedial work. This…

Whenever one uses ordinary least squares regression, one is making an implicit assumption that all of the independent variables have been measured without error. Such an assumption is obviously unrealistic for most social data. One approach for estimating such regression models is to measure implied coefficients between latent variables for which…

The advantages of the Delphi Technique are that it relies on collective expert judgment, prevents the excessive influence in decision-making of those in positions of status, and encourages consensus. The technique consists of five steps: (1) asking experts to list items relevant to a topic; (2) returning the list to the participants and asking…

A model permitting construction of algorithms for the polynomial conjoint analysis of similarities is presented. This model, which is based on concepts used in nonmetric scaling, permits one to obtain the best approximate solution. The concepts used to construct nonmetric scaling algorithms are reviewed. Finally, examples of algorithmic models for…

A variety of path models involving unmeasured variables are formulated in terms of Joreskog's general model for the analysis of covariance structures. (Author)

This memorandum is the manual for the VARAN (VARiance ANalysis) program, which is the latest addition to a series of computer programs for multivariate analysis of variance. As with earlier programs, analysis of variance, univariate and multivariate, is the main target of the program. Correlation analysis of all types is available with printout in…

Populations of factorially simple and complex data were generated with first the oblique and orthogonal factor models, and then solutions based on special cases of the general orthomax criterion were compared on the basis of these characteristics. The results are discussed and implications noted. (DEP)

Current practice in structural modeling of variables is limited to means and covariances based on multivariate normality assumptions. This article extends structural equation models to higher order product moments and to non-normal distributions. Areas of possible research are described. (Author/JKS)

A three-dimensional model made from pegboard, rods, and spheres can be used to illustrate principles of multi-variate analysis. (SD)

An example demonstrates that Joreskog's suggested sufficient conditions for identifying unknown parameters in a confirmatory factor analytic model with correlated factors are not sufficient. Sufficient conditions are presented. (Author/JKS)

Situations for which the researcher should use component analysis versus common factor analysis are discussed. Topics addressed include key algebraic similarities and differences, theoretical and practical issues, the factor indeterminacy issue, latent versus manifest variables, and differences between exploratory and confirmatory analysis…

Asymptotic distributions of the estimators of communalities are derived for the maximum likelihood method in factor analysis. It is shown that equating the asymptotic standard error of the communality estimate to the unique variance estimate is not correct for the unstandardized case. Monte Carlo simulations illustrate the study. (SLD)

Two algorithms that automatically select subsets of variables (PACE algorithm) and reference variables (Fabin estimators), respectively, used for the noniterative estimators are presented. The PACE algorithm is based on a nonsymmetric matrix sweep operator. A Monte Carlo experiment compares the relative performance of these estimators and others.…

Scoring of response vectors to give maximum test-retest correlation is investigated. A general method is given for finding the best scores, deriving them for the normal factor model, and showing that for a standard model for binary response it is easy to approximate the best scores. (SLD)

Principal component, image component, three types of factor score estimates, and one scale score method were compared over different levels of variables, saturations, sample sizes, variable to component ratios, and pattern rotations. There were virtually no overall differences among methods, with the average correlation between matched scores…