Developmental Psychology Copyright 1997 by the American Psychological Association, Inc. 1997, Vol, 33, No. 4, 657-668 0012-1649/97/$3.00 Gender-Specific Pathways Between Maternal Depressive Symptoms, Family Discord, and Adolescent Adjustment Patrick T. Davies and Michael Windle Research Institute on Addictions Relations among maternal depressive symptoms, family discord, and adolescent psychologicaladjust- ment were examined in a sample of 443 middle adolescents and their mothers. Histories of maternal depressive symptoms, gathered at 3 occasions with 6-month intervals, were related to subsequent adolescent reports of depressive symptoms, conduct problems, and academic difficulties for girls but not for boys. Mediational tests indicated that girls' greater vulnerability to family discord (e.g., marital discord, low family intimacy, parenting impairments) accounted for the impact of maternal depressive symptoms on their social and emotional adjustment. Analyses suggest that family discord is a strong mediator in the development of girls' conduct disturbances and a modest mediator of girls' depressive symptoms. Results are discussed within a framework that integrates interpersonal models of parental depressive symptoms with the gender intensification hypothesis. Although having a parent with high levels of depressive symp- toms probabilistically increases children's risk for a wide range of psychological problems (e.g., depression, conduct prob- lems), many, if not most, children develop along competent and adaptive developmental trajectories (Cummings & Davies, 1994b; Downey & Coyne, 1990). The heterogeneity in out- comes of these children has led to greater theoretical articulation of the processes that may modify (i.e., moderate)and account for (i.e., mediate) the link between parental depressive symp- tomatology and child adjustment (Cummings & Davies, 1994b; Dodge, 1990). Given that genetic and biological mechanisms only partly explain the transmission of psychopathology in de- pressed families, recent conceptual frameworks have increas- ingly emphasized the study of family and environmental risk processes as complementary directions for research. These inter- personal models have postulated that parental depressive symp- tomatology may indirectly affect children's psychological ad- justment through its association with family discord (Cum- mings & Davies, 1994b; Downey & Coyne, 1990; Hops, 1992). Thus, expressions of family discord characterized by conflict, distress, and low levels of intimacy and satisfaction are concep- tualized as proximal mediators in pathways between parent de- pressive symptoms and child psychological problems. The in- creasing interest in family disturbances as mediators of parent depressed mood has been paralleled by a focus on factors that may modify or moderate the effects of parental depression (e.g., Hops, 1995; Seifer, 1995). As part of this emphasis, emerging Patrick T Davies and Michael Windle, Research Institute on Addic- tions, Buffalo, New York. This research was supported by the National Institute on Alcohol Abuse and Alcoholism Grant AA07861. We are grateful to Levent Du- menci and Rebecca Windle for their comments on earlier versions of this article. Correspondence concerning this article should be addressed to Patrick T Davies, who is now at the Department of Clinical and Social Sciences in Psychology,University of Rochester, Rochester, New York 14627. developmental hypotheses have underscored the proposition that the nature and strength of mediational pathways may be con- strained by gender and age (see Hops, 1995). Accordingly, the present study had two primary aims. Focus- ing on the mediational model, the first aim was to test a model whereby dimensions of family discord accounted for the rela- tionship between history of maternal depressive symptomatol- ogy and child adjustment problems. Given that children's gender may modify the effects of depressive family contexts, the second aim was to investigate whether the nature of the mediational pathways differs for girls and boys during middle adoles- cence--a developmental period marked by intensification of gender roles (Hill & Lynch, 1983; Nolen-Hoeksema, 1994). At the outset, however, it is important to clarify that this study focuses on depressive symptomatology, a construct that is dis- tinct from clinical diagnoses of depressive disorder despite some conceptual overlap (Fergnsson, Horwood, & Lynskey, 1995). Although adolescents may be more developmentally mature than children, developmental risk models commonly stress that adolescence may be a period of heightened vulnerability to pa- rental dysphoria (Gelfand & Teti, 1990; LaRoche, 1989). For example, the stress of living with a distressed parent may delay the successful resolution of numerous stage-salient challenges that children face in middle adolescence (e.g., career decisions, independence from parents, establishment of intimate relation- ships), thereby resulting in increased risk for psychopathology (Gelfand & Teti, 1990; LaRoche, 1989). Parental depression may also undermine the successful resolution of these age-ap- propriate life tasks through poor role modeling (e.g., poor cop- ing skills), the provision of limited parental social support, and a demanding or hostile expressive style when interacting with their adolescent offspring. Relatedly, the peak occurrence of stressful life events in middle adolescence may accentuate the stress of living with a distressed parent (Windle, 1992), particu- larly given children's greater sensitivity to parental and family distress as they reach adolescence (Buchanan, Maccoby, & Dornbusch, 1991; Cummings & Davies, 1994b). Consistent 657