A multi-test alcoholic taxonomy: canonical coefficient clusters

Area of Science:

• Psychology
• Psychopathology
• Personality Science

Background:

• Subtyping alcoholic personalities has lacked convergence due to methodological limitations.
• Previous research often relied on single, theoretically limited, or distress-heavy inventories.

Purpose of the Study:

• To transcend idiosyncratic cluster solutions in alcoholic personality research.
• To establish a robust, theoretically grounded method for identifying alcoholic personality subtypes.

Main Methods:

• Utilized canonical correlation between the Basic Personality Inventory (BPI) and the Millon Clinical Multiaxial Inventory (MCMI).
• Cluster analyzed the resulting global canonical variates to derive a six-cluster solution.
• Described subtypes based on canonical variate scores and psychometric inventory data.

Main Results:

• Identified a six-cluster solution representing distinct alcoholic personality subtypes.
• Characterized these subtypes through their unique profiles on canonical variates and underlying psychometric measures.
• Demonstrated a more convergent approach to personality subtyping in alcoholism.

Conclusions:

• The proposed method offers a more stable and theoretically appealing approach to subtyping alcoholic personalities.
• The identified six subtypes provide a framework for understanding heterogeneity in individuals with alcohol use disorders.
• This research advances the field by providing a psychometrically sound basis for personality-based alcoholism subtypes.