In this study I assessed a causal model between caregivers’ prior relationship to care-receivers, commitment to the relationship, and depression in parental and spousal caregiving, based on social exchange and commitment theory. Data (N=695) from the National Survey of Families and Households (NSFH) were used to test a path model examining the effects of relationship quality and commitment, as well as ace, gender, income, education, health, living arrangement, emotional support, and adult children's marital status on depression. This study began the process of combining the social psychological concept of commitment and the gerontological caregiving literature. The expected effects of commitment on depression were not statistically significant for either spousal caregiving or parental caregiving. As for spousal caregiving, caregivers’ health and relationship quality were negatively associated with caregivers’ depression. In parental caregiving, caregivers’ education and health had negative effects on caregivers' depression. The quality of the relationship with spouse or parent was notable for explaining commitment to the relationship. The predicted positive effect of relationship quality with parent on moral commitment was contradicted by a statistically significant finding of a negative effect. Spousal caregivers’ structural commitment to marital relationship was positively affected by the quality of the relationship with spouse. / Ph. D.