Reise, Cook, and Moore proposed a "comparison modeling" approach to assess the distortion in item parameter estimates when a unidimensional item response theory (IRT) model is imposed on multidimensional data. Central to their approach is the comparison of item slope parameter estimates from a unidimensional IRT model (a restricted model), with…

A FORTRAN program for transforming raw scores into square roots, natural logarithms, arc sines, and other transformations is described. Any or all transformations may be made on a data matrix of fifteen columns or fewer, and one hundred rows or fewer. (Author/JKS)

Extended Lubin's discussion of ordinal and disordinal interactions by introducing the concept of dual-ordinal. This concept was considered important since a transformation cannot bring about additivity unless the interaction is dual-ordinal in nature. A method is presented for determining whether or not an interaction is dual-ordinal. (Author/BJG)

A general factor analysis computer algorithm is briefly discussed. The algorithm is highly transportable with minimum limitations on the number of observations. Both singular and non-singular data can be analyzed. (Author/JKS)

A rational approach to standard transformed scores which always results in the numerical to letter grade correspondence for any chosen hypothetical grade distribution model is developed. (Author/LMO)

The purpose of this paper is to outline the role of orthogonal rotation in canonical analysis, including the evaluative measures that need be reported and scrutinized upon application. (Author)

A Monte Carlo simulation was used to investigate the effect of data transformation on average correlations between Job Descriptive Index scores. In general, results indicate that transformation of scores did not result in marked changes in the correlations and that such transformation may not be worthwhile. (SLD)

Describes a program for the T-score technique of normal standardization. T-scores transform a raw score distribution, regardless of its skewness or kurtosis, into a normal distribution with a mean of 50 and a standard deviation of 10. (Author/RC)

Formulas for transforming scores and statistics on a rating scale having any number of categories to a scale having a different number of categories are described. To illustrate the use of one of the formulas, the means and variances of items on six forms of a course evaluation questionnaire were compared. (Author/LMO)

This method of generating multivariate data differs from previous techniques in that it uses Crout factorization to develop the desired variance-covariance matrix. (Author/CK)

Three factors that increase score variability yet can be associated with an increase, a decrease, or no change in Pearson's correlation coefficient (r) are discussed (restriction of range, errors of measurement, and linear transformations of data). The connection between changes in variability and r depends on how changes occur. (SLD)

A method requiring minimal computational effort is presented for transforming ordered residuals for purposes of testing the correctness of a regression model. The method maintains the same logical ordering in the transformed residuals as that of the original residuals and is suitable for either correlated or uncorrelated data. (Author/BS)

The use of the arc-sine transformation in analysis of variance can lead to difficult inference situations and pose problems in interpretation. It can also produce tests of noticeably lower power when the null hypothesis is false, and is not recommended as a standard tool. Simulated illustrations are provided. (Author/GDC)

The effects of under and overrotation on common factor loading stability under three levels of common variance and three levels or error are examined. Four representative factor matrices were selected. Results suggested that matrices which account for large amounts of common variance tend to have stable factor loadings. (Author/RL)