Intentional control over prejudice: when the choice of the measure matters

We tested the idea that the col relation between implicit and explicit meas ures of prejudice depends on whether or not groups ale normatively protecte d against discrimination. A pilot study (N = 31) showed that 13 categories varied widely in the degree to which if is acceptable to express negative o pinions about them. The main study involving 89 Catholic subjects found tha t explicit (reward allocation, liking ratings) and implicit measures (lingu istic intergroup bias) of prejudice were correlated for the out group that is not normatively protected against discrimination (Islamic Fundamentalist s) but uncorrelated for the outgroup that is protected (Je,vs). Copyright ( C) 1999 John Wiley & Sons, Ltd.