Segura and Fraga
17
smaller, not surprising since many of those are lost to the analysis when Republicans are
excluded. Nevertheless, Anglo Democrats and Independents are 16% more likely to have
defected from Bustamante than Latinos, and white Conservative Democrats and Independents
are 20% more likely to have defected than liberals.
What, then, can we conclude from the analyses presented in Tables 4 and 5? It is
difficult to assess motive with survey data. But the evidence regarding patterns of behavior is
provocative. Over 1/3 of all non-Hispanic white and Asian voters who supported Davis in the
recall chose not to vote for Bustamante. In addition, self-identified conservatives—even those
not in the GOP—were far more likely to defect, as were supporters of the racially charged Prop
54. So while we cannot say with certitude what share of defectors did so for explicitly racial
reasons, it appears highly likely that racial sentiment was responsible for some meaningful share
of the fall-off between opposition to the recall and support for Bustamante.
Turncoats—Bustamante Voters who Supported the Recall
The second curious pattern—with potentially racial overtones—that we wished to explore
was that of recall supporters who, surprisingly, voted for Cruz Bustamante in the replacement
election. Others have suggested that some liberal Democrats and Latinos, frustrated with Davis’
inaction on matters of concern to them, might have voted strategically by supporting the recall in
hopes of installing Bustamante in the governorship. What evidence is there for this claim?
As we described in Table 2, the overall numbers of folks engaging in this pattern of
behavior is small, from a low of .47% of Asians to a high of 7.2% of Latinos. Nevertheless, as
we can see in Table 3, Latinos—who presumably are best served by this strategy, should it have
succeeded—do, in fact, engage in this behavior with some frequency. Among Latino recall
supporters, 15.5% went on to support Bustamante. And surprisingly, African Americans did so
Convention