Numerous scholars of racial politics and political behavior have documented that
the racial context affects the political behavior of both whites and blacks. Drawing from
V.O. Key’s classic work on Southern politics to more recent research beyond the South,
whites seem to react negatively to the presence of blacks within close geographical
proximity.
Much of this work focuses on individuals’ political attitudes on racial
policies, prejudice and vote choice, motivated by a “power-threat” hypothesis which
predicts more negative attitudes as black group size increases (Giles and Evans 1985,
1986). Scholars, for example, find that high Black population concentrations in
Louisiana parishes are associated with greater Republican party identification (Giles and
Hertz 1994) and greater support for conservative segregationist David Duke's senatorial
candidacy (Giles and Buckner 1993; but also see Voss 1996 and Giles and Buckner 1996).
More broadly, however, Oliver and Wong (2003) find no evidence that a greater size of
out-group population is correlated with racial hostility. Rather, they demonstrate that a
greater size of in-group population within neighborhoods promotes negative attitudes
toward other racial and ethnic groups, while greater out-group population weakens them.
While the power-threat hypothesis has also motivated some research on the
effects of racial context on whites’ decisions to participate, the theoretical expectations in
predicting how individuals’ participation decisions respond to minority group size or
increasing racial diversity are not as strong. Leighley and Vedlitz (1999) argued that
Anglos would respond to minority group threat by engaging in political activity to protect
their advantaged position, an expectation one might argue is broadly in line with the
1
Others have investigated whether increased black presence leads to greater contact and
reduces prejudicial attitudes, resulting in more positive inter-group perceptions (see, for
example, Sigelman, Bledsoe, Welch and Combs 1996; Stein, Post, and Rinden 2000;
Welch, Sigelman, Bledsoe, Combs 2001).
1