When she brings home the bacon: Labor-force participation and the risk of spousal violence against women.

Because employment plays a central role in shaping gender identities and gender relations, it has important implications for understanding women's risks of spousal violence. This article analyzes the relationship between participation in the labor force and the risk of spousal violence against women by treating employment as a symbolic, rather than simply socioeconomic resource. We begin with a latent class analysis that identifies qualitatively distinct patterns of violence against wives. We then examine direct and conditional effects of employment on women's risks of spousal violence. Our results show that the effect of a woman's employment on her risk of spousal violence is conditioned by the employment status Of her partner To some extent, these effects reflect efforts by men to coercively control their female partners. Key Words: employment, spousal violence. One of the major structural transformations of North American society since World War 11 has been the movement of married women into the paid labor force. The consequences of this transformation for family life and for marital relations, in particular, have been the focus of considerable research (e.g., Blood & Wolfe, 1960; England & Farkas, 1986). One aspect of marital relations that has been somewhat neglected by this work is violence by husbands against wives. Although many agree that the employment statuses of wives and husbands are likely to affect the risk of marital violence (Allen & Straus, 1980; Johnson, 1995), empirical research on this relationship is inconclusive. Previous research on the relationship between employment and marital violence has largely focused on whether rates of wife battering vary by general socioeconomic status (Moore, 1997; Schwartz, 1988). From this perspective, employment is but one indicator of access to economic resources, and the explicit or assumed mechanism linking unemployment to violence is either the stress that a lack of resources places on marriages or women's economic dependency. Yet employment provides more than just economic resources. It has crucial symbolic importance for identities, selfesteem, and mental health (Gecas, 1989; Kohn & Schooler, 1983). For men, in particular, working is a critical means of constructing masculinity (Connell, 1995; Thoits, 1992). For this reason, the effects on marital violence of employment as a symbolic rather than a solely economic resource need to be considered. We do so by assessing the independent and combined effects of wives' and husbands' employment statuses on husbands' violence against their wives, controlling for other measures of socioeconomic status. This allows us to evaluate different theoretical perspectives on violence against wives, as well as shed light on the complex relationships among class, gender relations, and marital violence. Employment, Economic Resources, and Spousal Violence Employment statuses of husbands and wives are conventionally conceptualized as indicators of access to economic resources. From one perspective, diminished resources lead to stress, frustration, and conflict that can culminate in violence between spouses (Dutton, 1988; Gelles, 1974). Resource deprivation that accompanies unemployment should thus increase risks of spousal violence. Violence should be particularly prevalent in relationships in which both partners are unemployed, but employment of either spouse should lessen the risk of violence. If employment is simply an indicator of economic resources, however, it should have little effect on spousal violence, independent of more direct measures of economic resources, such as household income. From another perspective, it is women's employment that affects their risks of spousal violence. Kalmuss and Straus (1990) argue that employed women are less dependent on their husbands, whereas unemployed women lack the resources to leave abusive relationships or negotiate changes in their partners' behaviors. …