Married and recently divorced mothers' stressful events and distress : Tracing change across time

FREDERICK O. LORENZ, RONALD L. SIMONS, AND RAND D. CONGER Iowa State University GLEN H. ELDER, JR. University of North Carolins* CHRISTINE JOHNSON AND WEI CHAO Iowa State University** We examined the relationship between divorce and psychological distress in adult women by combining data from two studies, one of divorced mothers with adolescent children (n = 188) and another of married mothers (n = 306). The data were modeled using latent growth curves to chart change over time. Change in depressive symptoms paralleled changes in stressful events. For divorced mothers, stressful events and depressive symptoms increased significantly soon after the divorce and then diminished over the next 3 years, although not to the same levels reported by married women, The data were consistent with a proposed model that incorporated aspects of both the social causation and the selection perspectives. Key Words: depressive symptoms, divorce, latent growth curves, panel studies, stressful events. The dramatic increase in divorce rates in the United States is well documented. Current estimates project that at least half of recent first marriages will end in divorce (Bumpass, 1990). Despite its frequency and growing acceptance, divorce is still a traumatic life change; single parents have higher rates of both emotional and physical health problems than their married counterparts (Amato & Keith, 1991; Bachrach, 1975; Kitson, 1992; Tschann, Johnston, & Wallerstein, 1989). In this article, we examine the relationship between divorce and psychological distress by introducing stressful life events as a mediating link between family structure and depressive symptoms in women. This study is conducted by combining data from two three-wave panel studies, one of recently divorced mothers with adolescent children and another of a parallel sample of married mothers. Although several earlier studies have used panel data (see Kitson, 1992, pp. 24-25), this combined data set permits us to extend previous research in specific methodological and substantive ways. Methodologically, we extend previous panel studies on divorce and distress by using latent growth curves to explicitly model intra-individual change in psychological distress and stressful life events in recently divorced mothers, and we then compare their trajectories of change with those of the parallel sample of married mothers. Substantively, we draw on traditional research of family stress (McCubbin & Patterson, 1983) and the stress-distress process (Coyne & Downey, 1991; Kessler, Price, & Wortman, 1985; Mirowsky & Ross, 1989; Pearlin, 1989) to conceive of divorce as a primary stressor that makes women susceptible to an array of stressful life events. These events accumulate quickly following divorce and then subside over time. Changes in depressive symptoms parallel changes in stressful events. To accommodate the competing argument that some people experience more stressful events and higher levels of depressive symptoms because of past behavior problems and that these people are more likely to be among the divorced, we also incorporate measures of antisocial behavior into our model. Taken together, the mother's family structure and past antisocial behavior capture aspects of two competing perspectives: Social causation asserts that divorce creates conditions that make women susceptible to more stressful life events (e.g., Turner, Wheaton, & Lloyd, 1995) and, therefore, higher levels of distress; selection claims that women with a history of antisocial behavior are likely to experience more stressful events, become depressed, and are more likely to be among the divorced (e.g., Patterson & Dishion, 1988). THEORETICAL MODEL The connection between family structure and psychological distress is based on the empirical observation that both divorced men and women are more depressed than their married counterparts. …