Latinos, participation and *education.

Research in political science has highlighted the potential importance of factors such as race, gender and ethnicity for political participation, and use of intercepts to capture such differences has become common in quantitative models of participation. Because these models do not allow the coefficients of variables predicting participation to vary across groups, they are limited in their ability to explain group variations in the effect of resources, and not just the levels of resources, on participation. This dissertation illustrates limitations of such an approach through quantitative analysis of the effect of education on the political participation of whites and Hispanics. Using data from the American Citizen Participation Study of 1990 and the Latino National Political Survey of 1989--90, it examines a parsimonious model of participation, and asks why the effect of education on participation is substantially greater for whites than Hispanics, other things equal. It finds that the differential effects of education are not due to education's influence on one or two specific acts. It is unable to confirm whether a relative (rather than absolute) measure of education could explaining the disparity, although it does suggest that controls for the quality of education may be important. It also finds mixed support for the idea that education's disparate effect on Latinos and whites is due to differences in socialization, access to alternative sites of acquiring participatory resources, or the existence of a Latino elite who are unable to acquire educations without a preponderance of other resources. It finds somewhat stronger support for the idea that white-Latino differences in the impact of education on participation are due to education's role in sorting individuals into social strata in which participation is more or less important. It also finds that two primary justifications for distinguishing between Latinos and whites, differences in ethnicity between national-origin groups and immigration status, are critical for unraveling white-Latino differences in education's effect upon participation. The dissertation reveals that analyses that do not consider national-origin status or location of socialization will misrepresent the ways in which education works to influence participation. The dissertation concludes by outlining directions for future research.Ph.D.Bilingual educationEducationEthnic studiesPolitical scienceSocial SciencesUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/123663/2/3096164.pd