Which statistical method is used to classify observations into predefined groups based on predictor variables?

Which statistical method is used to classify observations into predefined groups based on predictor variables?

• A. Deviation contrast
• B. Discriminant analysis
• C. Direct oblimin
• D. DFBeta

Answer: (B)

Discriminant analysis is used to classify observations into predefined groups based on predictor variables. It works by finding linear (or quadratic) combinations of the predictors that maximize separation between the known groups, producing discriminant scores for each case. Each observation is then assigned to the group with the highest likelihood or closest score, making this a supervised classification method that uses group labels to build the model and apply it to new data. This contrasts with deviation contrast, which is about coding categorical predictors for ANOVA rather than classifying cases; direct oblimin, a rotation method in factor analysis; and DFBeta, a regression diagnostic measure of a case’s influence on coefficients.

Discriminant analysis is used to classify observations into predefined groups based on predictor variables. It works by finding linear (or quadratic) combinations of the predictors that maximize separation between the known groups, producing discriminant scores for each case. Each observation is then assigned to the group with the highest likelihood or closest score, making this a supervised classification method that uses group labels to build the model and apply it to new data. This contrasts with deviation contrast, which is about coding categorical predictors for ANOVA rather than classifying cases; direct oblimin, a rotation method in factor analysis; and DFBeta, a regression diagnostic measure of a case’s influence on coefficients.