W
EAVER
5
The Color of the Campaign
is somewhat arbitrary and unrealistic (voters usually know the race of both candidates).
Furthermore, the impact of skin color gradations on candidate evaluation has been ignored, even
while other studies have demonstrated that skin color is an important qualifier to racial
categorization.
Nayda Terkildsen’s experiment gave subjects only one fictitious candidate to evaluate.
There are obvious drawbacks to this, namely, that the scale used to evaluate candidates varies by
subject so they aren’t necessarily comparable across subjects. The study found that despite
identical issues stances, personal characteristics, and qualification white voters were less likely to
vote for a black candidate. Another experiment conducted on 656 whites did use 2 candidates
but voters were only asked to assess one candidate.
The study revealed that black
conservatives were viewed as more liberal than their white counterparts and black liberals were
more likely to be seen as incompetent. Black candidates were believed to be more
compassionate than whites.
In addition, more work is needed to delineate the “motivational basis” for raciallized
voting.
Previous analyses have shown that even if a subject supports minority issues
(affirmative action, bussing, set-asides), they are still not likely to cast a ballot for a black
candidate.
This study hopes to improve on the limitations of past work and data constraints by
utilizing a new methodology that combines the benefits of public opinion survey research with
new experimental analysis on voter psychology. This study will extend the experimental
13
Another experimental study exposed subjects to two candidates but subjects were only asked to assess one of them. See
Sigelman, et al.
14
Terkildsen.
15
Sigelman, et al.
Convention