The utility of a nonlinear transformation of the criterion variable in multiple regression analysis is established. A well-known law--the Pythagorean Theorem--illustrates the point. (Author/JKS)

Three different uses of ratio variables in aggregate data analysis are discussed: (1) as measures of theoretical concepts, (2) as a means to control an extraneous factor, and (3) as a correction for heteroscedasticity. Alternatives to ratios for each of these cases are discussed and evaluated. (Author/JKS)

Equations for estimating the value of the multiple correlation coefficient in the population underlying a sample and the value of the population validity coefficient of a sample regression equation were investigated. Results indicated that cross-validation may no longer be necessary for certain purposes. (Author/MH)

Factor scale scores are sometimes used as weights to create composite variables representing the variables included in a factor analysis. If these composite variables are then used to predict some dependent variable, serious theoretical and methodological problems arise. This paper explores these problems and suggests strategies for circumventing…

Five category scores of word usage frequency were generated from responses on the Rotter Incomplete Sentences Blanks for 61 college freshmen. These scores were then used to predict asocial behavior of the freshman. (Author/JKS)

Sociological and demographic research often uses variables computed as ratios. When the denominators are highly correlated, and the ratios are used in correlation or regression analysis, a statistical dependency is formed. This article investigates this problem, particularly with respect to partial correlation, multiple regression, and the…

Background characteristics, school performance, and achievement test data were analyzed for 788 third-grade boys and 774 third-grade girls who were known later to have become high school dropouts or graduates. As early as the third grade a variety of variables were predictive of later status. (Author/JKS)

Computer-based Monte Carlo methods compared the predictive accuracy upon replication of regression of five complete and four incomplete factor score estimation methods. Prediction on incomplete factor scores showed better double cross-validated prediction accuracy than on complete scores. The unique unit-weighted factor score was superior among…

The inclusion of unmeasured variables in path analyses in educational research is discussed. The statistical basis for inclusion is presented, along with several examples. (JKS)

The relative magnitudes of R-squared values computed through multiple regression models using grade equivalent scores, raw scores, standard scores, and percentiles as both predictor and criterion variables are compared. Grade equivalents and standard scores produced the highest R-squared values. (Author/JKS)