Dependency Autonomy Stress and Depression: A Test of an Interactive Model

Two diathesis-stress models were used to predict depression. The additive model combined dependency and autonomy (dependency-autonomy) as the diathesis and also combined dependent and autonomous hassles or life events as stressors. The congruent model used either dependency and dependent stressors or autonomy and autonomous stressors. Ninety-seven female and 42 male undergraduate students completed self-report measures of dependency and autonomy, hassles and life events, and depression at three testing sessions one week apart. Hierarchical multiple regressions — with prior depression entered first and followed by dependency and/or autonomy, stress, and their interaction — indicated that additive models with a significant interaction accounted for more variance in predicting depression than did the only congruent model that produced a significant interaction. These results suggest that dependency and autonomy may be non-orthogonal and even components of the same larger construct. Tentative empirical evidence was also found to support Beck's (1983) notion that an individual's predominant personality mode may change over time.