Traumatic Events and Marital Well-Being

Abstract

Using a national probability sample, we investigate the impact of traumatic events on marital well-being. We argue that traumatic life events such as the death of a child, a life-threatening illness, and physical attack have a negative impact on marital well-being and that this impact is moderated by spousal support. Partially confirming our first hypothesis, we found that physical attack is a significant predictor of marital wellbeing. Respondents who had been physically attacked in the past reported lower levels of marital satisfaction. As predicted, spousal support critically affected the relationship between traumatic events and marital well-being. Key Words: marital quality, marital satisfaction, spousal support, traumatic events. Although several researchers have explored the correlates of marital satisfaction (Broman, 1991; 1993; Crohan & Veroff, 1989; Glenn, 1989; Glenn & McLanahan, 1982; Glenn & Weaver, 1978; 1988; McLanahan & Adams, 1987; Miller, 1976; White, 1983; White, Booth, & Edwards, 1986), much of the research does not address the role of life stress. In this research, we focus on how the role of traumatic life events affects marital satisfaction and happiness. Consistent with the literature on stress, we advance and test two predictions: that traumatic life events have a negative impact on the sense of satisfaction in marriage and that this effect is moderated by spousal support. THE LIFE COURSE PERSPECTIVE We use a life course perspective as an overarching framework for our discussion of traumatic life events. We argue for the merging of a life course perspective with life stress research as a means to understand the complexity of the impact of traumatic events on the marital relationship. Although the life course perspective and stress research are represented by vast bodies of literature, we have identified themes from the literature that are relevant to our present study. We begin by arguing for a broader conception of as it relates to the life course. Most studies that utilize a life course perspective focus on normative events and transitions such as role occupation (Moen, Dempster-McClain, & Williams, 1989; 1992), employment patterns (McLanahan & Sorenson, 1985), parenthood (Umberson & Gove, 1989), and changes in household composition (McLanahan & Sorenson 1985). By normative, we mean that certain events in a family's life course can be anticipated to one degree or another. For example, we expect to see fluctuations in financial security within a family as it goes through various stages in the life course. Therefore, families save money in times of relative prosperity in anticipation for a rainy day. This idea of normativeness is in contrast to our use of traumatic events, which are uncontrollable and extremely negative. In doing this, we expand the operationalized domain of the life course perspective by focusing on the kinds of life events-like death of a child, criminal victimization, or suffering from a life-threatening illness-that are conventionally investigated in stress research. Even when the focus of life course research is on undesirable events (Conger, Lorenz, Elder, Simons & Ge, 1993), there is still the sense that these events are part of an expected norm of family life course trajectories (such as parent and child conflict, divorce or separation). Traumatic events, which cannot be anticipated, are generally beyond the control of individual family members, and, therefore, are not normative. The second theme we build on is that the past can have very real consequences in the present. We look at the relationship between suffering a traumatic event and its impact on the present marital relationship. Again, this extends the notion of life event because we are dealing with uncontrollable, extremely negative life events that generally are not considered normative. Much of the research that examines the present impact of past events retains a normative definition of events and focuses on issues such as role overload and family conflict (for example, Moen et al. …