The Psychology of Prejudice

condition (r = -.52; p < .05) but near zero in the covert condition (r = -.08). For the quest dimension, the correlation is near zero in the overt condition (r = -.10) but significantly negative in the covert condition (r = -.45; p < .05).³

As in the Batson et al. ( 1978) study, then, the intrinsic, end dimension correlates negatively with an overt measure of racial prejudice. But this significant negative correlation disappears when choosing to sit with the White person could masquerade as a movie preference. The pattern of behavior associated with devout, intrinsic religion in these two studies is strikingly reminiscent of what Gaertner and Dovidio ( 1986) called aversive racism. The pattern suggests a concern on the part of the intrinsically religious not to be seen by others, or by themselves, as prejudiced; still, "the underlying negative portions of their attitudes are expressed, but in subtle, rationalizable ways" ( Gaertner & Dovidio, 1986, p. 62). Only the quest dimension has a significant negative correlation with the covert measure of prejudice.

We have traveled a long and winding path in this review of the religion-prejudice relationship, but we believe that we have gotten somewhere. First, we found much research suggesting a positive correlation between being religious and being prejudiced. Then we found much research suggesting that this was only

-165-