rather than compulsive. Equality of opportunities is largely accepted, casual relations are widely tolerated, but intimate relationships with colored people are frowned upon. Mulattoes are generally less discriminated against than Negroes, but a small minority "prefers" Negroes to mulattoes. This small minority exhibits a much more virulent form of prejudice against both Negroes and mulattoes than does the general sample. Sex is an important determinant of prejudice. So is socioeconomic status, although our data are too uncertain and incomplete to determine the exact relationship. Ethnic origin of the parents likewise plays an important role.
The weaknesses of our study are many and obvious. As we have pointed out, the sample is not random nor proportional; the postulates underlying the analysis are debatable; and so on. Our conclusions must be accepted with all caution, and we have raised more problems than we have solved. Although our findings largely confirm previous studies, certain revisions of the literature seem in order. Should our study only stimulate criticism, further research, and a few working hypotheses, we should be highly satisfied.
The same terms such as "Negro" and "white" may, in different societies, cover objectively dissimilar groupings as exemplified by Brazil and the United States. In this research, we shall use the Brazilian definition. "Racial prejudice" is the totality of reciprocal relations of stereotypy, discrimination, and segregation existing between human groupings that consider themselves and each other as "races."