- The cross-sectional analyses show that as the proportion of one racial group in a classroom increases, other students become friendlier toward that group.
- The longitudinal analyses show that as the Black proportion decreases, Blacks are more likely to change from a weak to a stronger friendship tie. This effect is much weaker for Whites.
- Racially balanced classrooms maximize the interracial friendships for Blacks and Whites.
- Results provide strong support for the opportunity hypothesis and show little evidence that being in the racial minority diminishes interracial friendliness.
- The proportion of cross-race and same-race best-friend choices is greater for blacks than for whites, reflecting the greater friendliness of Blacks.
- Proportion Black has a statistically significant negative ef- fect on the likelihood that a Black student will choose a White classmate as best friend in five of the six time points over the school year.
- Black students are less likely to choose Whites as best friends as the proportion of Blacks in the class increases and are more likely to choose Whites as the proportion of Whites increases.
- Apparently neither Blacks nor Whites feel so threatened by being in the minority that they insulate themselves from the other race and form closer same-race friendship ties than they would in segregated classrooms.