Tests were constructed to measure three dimensions of prejudice related to interracial communication: stereotyping, affective reactions, and communicative social distance. A test was also constructed to measure the type of attributional choices made about blacks. The attributional choices were either defensive (negative and stereotypical) or nondefensive (positive and nonstereotypical). The reliability of these tests was assessed and the following hypotheses were tested: (1) The attributions made about blacks are significantly different from the attributions made about people whose race is unknown. (2) Highly prejudiced people are significantly more defensive in their attributions about blacks than are randomly selected people in their attributions about actors whose race is unknown. (3) There is a positive correlation between each of the three dimensions of prejudice on attributional defensiveness. (4) The effects of the three dimensions of prejudice on attributional defensiveness are additive such that each dimension has a significant, unique, and independent contribution to attributional defensiveness. Analyses of responses from 300 randomly selected college students confirmed the four hypotheses. (Author/RL)