First in a two-part series
Despite revolutionary changes in the 20th century, race relations still produce some of America's biggest challenges and dramatic issues. Public opinion figures provide ample evidence of positive social change, but also of differing views on how much things have improved. For example, a June 2003 poll showed that 87% of American adults feel that over their lifetimes, civil rights have greatly or somewhat improved the situation for blacks in the United States. However, that figure drops to 71% among blacks alone.
But what about the future? Can poll data offer any kind of leading indicator about the quality of race relations? One approach to this task is to gather information from young people. The Gallup Youth Survey has tracked teens' opinions about race-related matters for years. But simply investigating teens' level of exposure to members of other races -- by asking directly about the racial mix in their peer groups, schools, and neighborhoods -- may provide another useful angle on teens' perspectives.
Circle of Friends
The latest Gallup Youth Survey* did just that. Teens were first asked how many of their circle of friends -- the ones they hang out with on a regular basis -- are members of a different racial or ethnic group. Overall, half said "a few," and another 17% said "a lot."
Regional breakdowns reveal some interesting differences. Teens living in the South and West are more likely than those in the Northeastern and Midwestern regions to say they have a lot of friends of other races -- perhaps reflective of the large minority populations in those regions. That's particularly the case in the western states, where Hispanics account for 26% of the population, according to 2002 Census estimates.
In an attempt to further gauge teens' opportunities to interact with members of other races, Gallup asked respondents about the racial composition of their neighborhoods. Nationwide, 6 in 10 teens said their neighborhoods are all or mostly white.
Again, regional breaks turn up some informative differences. Despite the fact that the Northeast has a smaller percentage of whites overall than the Midwest (the 2002 Census puts the figures at about 72% vs. 81%, respectively), fewer white teens in the Northeast say their neighborhoods contain any nonwhites, reflecting less residential integration in the northeastern states. In the West, the number of teens who say their neighborhood is all or mostly white is very low even relative to the low overall white population (58%) in that region.
What about the schools? Both the overall and regional numbers reflect a higher level of integration in the schools than teens are exposed to in their neighborhoods, though some basic regional differences remain. Overall, 46% of teens said the students in their schools are all or mostly white, 31% said their schools are about half white and half nonwhite, and 18% said that students in their schools are all or mostly nonwhite. Teens in the South and West are less likely than those in the Northeast and Midwest to say their schools are homogenous.
Overall, these regional breaks suggest that adolescents' exposure to different racial groups in the different spheres of their lives is related to the population distribution, although in some regions -- particularly the western states -- integration of teens with members of different races appears to be higher than in others, even relative to the minority population in their area.
Across regions, teens seem to be more likely to be exposed to members of other races at school than in their home lives. The second part of this series will examine students' perceptions of interracial friendships within their schools and compare responses of whites and nonwhites to the questions above.
*The Gallup Youth Survey is conducted via an Internet methodology provided by Knowledge Networks, using an online research panel that is designed to be representative of the entire U.S. population. The current questionnaire was completed by 517 respondents, aged 13 to 17, between Aug. 1 and Aug. 29, 2003. For results based on the total sample, one can say with 95% confidence that the maximum margin of sampling error is ±5 percentage points.