Ridge regression is an approach to ameliorating the problem of large standard errors of regression estimates when predictor variables are highly intercorrelated. An interactive computer program is presented which allows for investigation of the effects of using various ridge regression adjustment values. (JKS)

Ridge regression is an approach to the problem of large standard errors of regression estimates of intercorrelated regressors. The effect of ridge regression on the estimated squared multiple correlation coefficient is discussed and illustrated. (JKS)

Ridge regression is a technique used to ameliorate the problem of highly correlated independent variables in multiple regression analysis. This paper explains the fundamentals of ridge regression and illustrates its use. (JKS)

The use of interaction and non-linear terms in multiple regression poses problems for determining parsimonious models. Several computer programs for using these terms are discussed. (JKS)

The type I error rate in stepwise regression analysis deserves serious consideration by researchers. The problem-wide error rate is the probability of selecting any variable when all variables have population regression weights of zero. Appropriate significance tests are presented and a Monte Carlo experiment is described. (Author/CTM)

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)

A suppressor variable is a regressor in a multiple regression which contributes more to the squared multiple correlation than the magnitude of its simple correlation with the outcome variable. An example of such a situation is provided for teaching purposes. (JKS)

This paper discusses how full model dummy variables can be used with partial correlation or multiple regression procedures to compute matrices of pooled within-group correlations. (Author/CTM)

The construction and interpretation of confidence intervals for the prediction of new cases in multiple regression analysis is explained. An example is provided. (JKS)

The use of multiple regression analysis was compared to the use of discriminant function analysis in the prediction of college faculty rank. The multiple regression technique was shown to be generally superior in this instance. (JKS)

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)

The three-year impact of a remedial arithmetic program on eligible St. Louis Public School pupils was investigated. Hypotheses were tested through multiple linear regression models for analyses of covariance. No treatment effects were found. The study reveals that changes in future program evaluation designs are needed. (Author/JKS)