- Friendship segregation peaks in moderately heterogeneous schools but declines at the highest heterogeneity levels.
- The generally positive relation between heterogeneity and friendship segregation suggests that integration strategies built on concentrating minorities in large schools may accentuate friendship segregation.
- Friendship in the Add Health sample are highly segregated by race. In general, the average value of alpha over the entire sample is 1.84, indicating that the average odds of nominating a same-race friend are about 1.8 times the odds of nominating a cross-race friend.
- Much of the observed friendship segregation in schools is due to factors such as belonging to the same clubs, having similar behaviors, and maintaining social balance.
- Results suggest that while friendship segregation tends to increase with racial heterogeneity, it need not, as some settings exhibit near perfect friendship integration, even at high levels of racial heterogeneity.
- Findings suggest that increasing racial diversity within the school might heighten relational segregation. As heterogeneity increases in a school, opportunities for making cross-race ties increase, but the observed rate at which cross-race ties are made does not keep up with that opportunity.
- The three focal organization characteristics (proportion in nonacademic tracks, grade segregation, and extracurricular mixing) each show some relation to segregation.
- Friendship segregation is lower in schools with integrated extracurricular programs. The effect is strong.
- The Add Health data suggest that while concentration of minority students may increase the possibility for cross-group contact, revealed same-race friendship preference increases unless organizational steps are taken to mitigate self-selection into same-race friendship groups. Without organizational structures to ensure status equality within the school, minority concentration may lead to greater friendship segregation, even as the absolute number of cross-race ties increases.
- Findings suggest that while schools in the northeast may be more racially isolated, schools into eh south tend to exhibit the greatest level of friendship segregation, net of school heterogeneity.
- Findings reinforce the idea that the simple degree of racial mixing is not sufficient to understand substantive student racial interaction. As such, schools in the south start from a higher base of segregation that will be that much more difficult to overcome.
- Interracial mixing within settings that capture the positive elements of contact theory (status equality and interdependent action) promotes friendship integration.
- The problem of racial friendship segregation is inherently multilevel: individuals choose friends but do so within the opportunities and constraints provided by the school context.
- The combination of low overall presence and differential mixing patters results in decreased racial segregation at the highest level of heterogeneity.
- School organization affects racial friendship segregation by structuring interracial contact. The strongest effect of school organization on racial friendship is through extracurricular mixing.
- Schools could increase interracial contact by limiting cross-grade contact. Tracking, on the other hand, combines both exposure and contact features. When schools disproportionately assign minority to nonacademic tracks, they add an additional status distinction between races, lowering contact opportunity and decreasing cross-race friendship.