Former Ku Klux Klansman David Duke ran three statewide campaigns in Lo
uisiana. This note presents a GLS analysis of his voting support. It s
uggests that whites living in racially heterogeneous parishes were no
more likely to support Duke than those in less-diverse locales, contra
ry to the ''racial threat hypothesis.'' Suburban whites were particula
rly likely to back Duke. Giles and Buckner (1993) present contrary res
ults because their analysis contained measurement error and omitted va
riable bias, aggregated data for urban areas inappropriately, and used
OLS on grouped data.