Recent research has demonstrated that individuals experience negative
affect (e.g. discomfort) when they are more prejudiced than what their
personal standards allow. This affective response has been attributed
to the discrepancy between personal standards and prejudiced response
s, with discrepancy operationalized as the sum of difference scores be
tween a measure of personal standards and a measure of proneness to be
prejudiced. The use of an algebraic difference index to operationaliz
e discrepancy, however, is appropriate ii and only if the assumptions
that underlie an algebraic difference index are met. The present study
shows that these assumptions are not met. The implications of these f
indings for research on prejudiced responses and affect are discussed.
Recommendations regarding the use of alternative approaches for study
ing prejudice-related discrepancies are also offered.