Proposes the use of the group overlap concept as a basis for determining effect size. Group overlap may be assessed via prediction of group assignment, using predictive discriminant analysis. The effect-size index proposed is that of improvement-over-chance "(I)" classification. Makes some suggestions for cutoffs of "I" values…

The purpose of this study was to determine which of six methods of ordering variables in a discriminant analysis yields subsets of variables that have the greatest discriminatory power. One method is based on univariate mean-square (or F) ratios, a second method on stepwise ordering, two methods on linear discriminant function (LDF) variable…

The process of discriminant analysis is reviewed in four different aspects: focusing on formulations, interpretations, and uses of discrimination, estimation and classification. Other issues, problems, and recent developments in discriminant analysis are also reviewed. (Author/DEP)

Classification is a procedure through which individuals are classified as being members of a particular group based on a variety of independent variables. Two methods of makin such classifications are discussed; the quadratic method is seen to be superior to the linear under certain constraints. (JKS)

An alternative analysis of the two-group single response variable design is proposed. It involves the classification of experimental units to populations represented by the two groups. Three real data sets are provided to illustrate the utility of the classification analysis. A table of sample sizes required for the analysis is presented.…

Three methods of transforming unordered categorical response variables are described: (1) analysis using dummy variables; (2) eigenanalysis of frequency patterns scaled relative to within-groups variance; (3) categorical variables analyzed separately with scale values generated so that the grouping variable and the categorical variable are…

There are two distinct but related purposes for carrying out a "discriminant analysis": (1) discrimination, and (2) classification. The primary objective of this paper was to review the outputs of selected computer programs often used to carry out a "discriminant analysis" with respect to these two purposes. Information provided by the programs on…

Using computer simulation data, the effect of a single global outlier in two-group classification analysis was explored in terms of the outcome variables of change in classification results (PCHNG), change in misclassification rate (MISDIF), and change in precision of misclassification rate estimation. The precision of misclassification rate…

Formulas for estimating cross-validated hit-rates, the number of correct classifications into an a priori grouping structure, were examined. The following mathematical formulas were compared: McLachlan's formula estimator, two Snappin and Knoke smoothed formula estimators, and the analytic leave-one-out estimator. The R method was included as a…

This study involved two phases: first when classification was based on the calibration sample, and second in a cross-validation setting. Computer-generated data were used. Results obtained from rules based on probabilities of group membership were compared for accuracy when classifying in the discriminant space and in the predictor variable…

Three estimates of the probabilities of correct classification in predictive discriminant analysis were computed using mathematical formulas, resubstitution, and external analyses: (1) optimal hit rate; (2) actual hit rate; and (3) expected actual hit rate. Methods were compared using Monte Carlo sampling from two data sets. (Author/GDC)

Eight discriminators were identified and data were obtained from the records of 80 graduate students who attained one of four achievement levels at the conclusion of a beginning course in educational statistics. Although the internal discriminatory power of the set of eight measures was very high, estimates of the true power were discouragingly…

A method is presented for extending full versus restricted model testing to predictive discriminant analysis when the criterion of model accuracy is classification hit rate. An example is presented involving the dropout or school persistence of 76 5th graders, and a computer program for the analysis is introduced. (SLD)

Linear classification functions (LCFs) arise in a predictive discriminant analysis for the purpose of classifying experimental units into criterion groups. The relative contribution of the response variables to classification accuracy may be based on LCF-variable correlations for each group. It is proved that, if the raw response measures are…

A linear classification rule (used with equal covariance matrices) was contrasted with a quadratic rule (used with unequal covariance matrices) for accuracy of internal and external classification. The comparisons were made for seven situations which resulted from combining three data conditions (equal and unequal covariance matrices, minimal and…