Most individuals have general political and attitudes toward government involvement in family matters (Erikson, Luttbeg, & Tedin, 1991; Sears & Citrin, 1985). Therefore, it is reasonable to expect that child care is an issue on which the public might express its general political attitudes. Moreover, one might expect public opinions on child-care issues to have some impact on the decisions of policy makers, although the general impact of public opinion on policy outcomes is open to debate (Becker & Heaton, 1967; Erikson et al., 1991; Miller & Stokes, 1963; Monroe, 1979; Page & Shapiro, 1983; Watson, Juster, & Johnson, 1991). Insofar as public opinion is expected to have at least some impact on the decisions of policy makers, public opinion on government involvement in child care is a phenomenon worthy of explanation. Despite the attention paid to childcare policy in recent years, there has been little systematic examination of the determinants of citizens' views toward government involvement in child care. Marttila and Kiley's (1988) research on attitudes toward child care and the Act for Better Child Care (ABC) addressed the issue of public opinion on government spending for child care. The study was based on a nationally representative sample of 901 respondents. In addition to the specific support expressed for the ABC by most respondents, nearly three fourths of respondents believed that the government should help families with child-care costs (Marttila & Kiley, 1988). The broad question remains: what explains variation in individuals' attitudes toward government spending on child care? To explore the determinants of opinions toward spending on child care, we developed and tested a theoretical model of public opinion towards government spending on child care. Specifically, we considered individuals' opinions about government spending on child care, as well as the factors that influence these opinions. Relying on the literature on individuals' political attitudes as a guide for variable classification, we also empirically tested two competing theories that might account for the development of and opinions about the role of government in child-care issues. THEORETICAL FRAMEWORK We considered two theoretical perspectives that have been important in the study of public opinion. The first, the symbolic politics perspective, depicts public opinion as a function of pre. dispositions and attachments developed during the socialization process. The second, the self-interest perspective, suggests that individuals' opinions on policy issues are structured by a rational comparison of the personal or familial benefits and costs that are associated with alternative policy viewpoints. In other words, individuals' attitudes are shaped by the self-interest that they associate with competing policy positions on a given issue. Symbolic Politics The basic premise of symbolic politics theory is that individuals make choices based on socialized predispositions acquired earlier in life (Citrin & Green, 1990; Lau, Brown, & Sears, 1978; Sears, Hensler, & Speer, 1979; Sears, Lau, Tyler, & Allen, 1980). One implication of this theory is that individuals respond to policies based on a symbolic association between the issue and their values (Citrin & Green, 1990; Sears et al., 1979). When faced with complicated decisions, they are inclined to make decisions that are consistent with well established values. There are three principles of this theory relevant for the current study. First, preferences held by most people are a result of childhood socialization and are maintained in adulthood (Citrin & Green, 1990; Lau et al., 1978; Sears et al., 1979). Second, attitudes are paired with positive and negative associations (Lau et al., 1978). Responses based on pre-adult socialization are triggered later in life by political symbols that resemble those earlier longstanding attitudes Sears & Citrin, 1985). …