Republican challenges to immigrants and Latinos in the mid-1990s. It is, perhaps, more
surprising that this effect is not seen among the naturalized citizens.
Second, the factors that shape the vote choice of the third and beyond generation are
richer than for those of the first and second generations. Issue evaluations did not prove to be
significant predictors of vote choice for the first and second generation. For the third generation,
issue evaluations supplemented candidate evaluations and partisanship in determining which
candidate to support (issues also played a greater role in shaping the partisanship of the third
generation). In the specification of the model with political party, a domestic and an
international issue distinguished Bush and Kerry supporters – the situation in Iraq and education
(education only proved significant in the specification without party). These two variables
proved to have more of an impact on vote choice than respondent evaluations of the candidates
or partisanship. As would be expected from previous scholarship on Latino issue preferences,
third-generation respondent evaluations of who would do a better job on education, among those
respondents who reported education to be the most important issue, proved to be the variable
with the most substantive impact on vote choice in the model.
Finally, controlling for these factors, the third generation is different in terms of vote
choice: it was more likely to support John Kerry. This suggests the foundation for future change
in Mexican American and Latino electorates. Many scholars (including, at least one of the
authors of this paper) have assumed that the stronger preference in the immigrant generation for
the Democrats necessarily translates into higher support for Democratic candidates. The
findings presented here should offer some caution to such assumptions. We find that immigrants
are, in fact, more likely to be Democrats. In terms of candidate choice, however, whether
controlling for partisanship or not, the naturalized citizen respondents were no more or less likely
to support John Kerry in 2004 when the other factors in the model are accounted for. This
39
Convention