What is a discriminant function variate?

What is a discriminant function variate?

• A. The differences between group means on the original variables.
• B. A score for an individual case on a discriminant function variate obtained by substituting that case's scores into the variate equation.
• C. The correlation between two measured variables.
• D. The eigenvalue associated with the discriminant function.

Answer: (B)

In discriminant analysis, a discriminant function is a linear combination of the predictor variables designed to separate predefined groups. The discriminant function variate is the actual score you get for a case when you substitute that case’s predictor values into the discriminant function equation, producing D = b0 + b1X1 + ... + bkXk. This score summarizes the case’s position along the discriminant dimension and is used to classify the case (for example, by comparing to a cutoff or using posterior probabilities). The idea isn’t just differences between group means or simple correlations; those describe data features or relationships, not the single score produced by the discriminant function. The eigenvalue relates to the amount of separation the function captures, but the variate itself is the computed score from the function.

In discriminant analysis, a discriminant function is a linear combination of the predictor variables designed to separate predefined groups. The discriminant function variate is the actual score you get for a case when you substitute that case’s predictor values into the discriminant function equation, producing D = b0 + b1X1 + ... + bkXk. This score summarizes the case’s position along the discriminant dimension and is used to classify the case (for example, by comparing to a cutoff or using posterior probabilities). The idea isn’t just differences between group means or simple correlations; those describe data features or relationships, not the single score produced by the discriminant function. The eigenvalue relates to the amount of separation the function captures, but the variate itself is the computed score from the function.