Grace Kao is fascinated with friendship. The Penn sociology associate professor has spent much of her 10 years here pondering what it means to have a friend, who befriends whom and whether having a best friend translates into better grades in the middle and high school years.

“Friendships are a central form of social support that are extremely important, especially during adolescence,” says Kao, who is using a data set of 90,000 children from about 130 schools nationally (collected by the National Institute of Child Health and Human Development in the 1990s) to learn more.

The data that Kao and graduate student Elizabeth Vaquera are interested in has to do with reciprocal friendships, and particularly interracial friendships. Children in the study were asked to nominate five same-sex and five opposite-sex friends. If two children said they were the other’s best friend, the friendship was called reciprocal. Kao says asking the question that way yields more accurate and useful results than asking the children if they have, for example, black friends or Asian friends. “We’re not doing that,” she says, “we’re asking them who their friends are.”
The researchers were curious about how different contexts might influence the quality and closeness of friendships, and whether reciprocity had any effect on academic outcomes.

What they discovered was “a little depressing,” says Kao, whose most recent paper, with Vaquera, “Do You Like Me as Much as I Like You? Friendship Reciprocity and Its Effects on School Outcomes among Adolescents,” will soon be published in Social Science Research.
The first thing Kao documented was friendship patterns. “For example,” she says, “are Asian Americans more likely to have white friends than African Americans? Are these friendships different?”

“The short answer is ‘yes,’” says Kao.

The researchers found striking differences by gender and race. Girls, it turns out, are substantially more likely­—a little over 60 percent versus a little over 50 percent—to have reciprocal friends than boys. Whites are more likely to have reciprocal relationships than African Americans, who are the least likely, and Asians fall somewhere in between. Native born youth also tend to have more reciprocal friendships than immigrant youth.
In sum, says Kao, “African American boys are the worst off. White girls are the best off. Here’s another way that minority kids are disadvantaged.”
Why minority children have fewer reciprocal friendships in school is something of a mystery. “From the data we can’t say a whole lot about why this happens,” says Kao. “From what else we know, it’s possible that African Americans’ closest friends are not in school but in the neighborhood.”

The vast majority of children choose friends of the same race, she says, and that becomes more pronounced as the children grow older. The more children of different races there are in a school, the more likely those children are to make friends of different races. “Intuitively it makes sense,” says Kao.

More surprising, she says, is how directly friendships can affect school outcomes. People who enjoy reciprocal friendships are more likely to feel part of the school and to feel that they belong, says Kao. And they also consistently get better grades.

The research clearly shows that social ties are important. People who have close friends also have social capital, and whether or not they are aware of it, the reciprocity factor matters, says Kao. In a related paper on the characteristics of interracial and intraracial friendships Kao focused on activities that children do together. She found that even among best friends, those who are interracial do fewer activities together during the week. In romantic relationships, the findings were similar. “In terms of what they do in private, they’re similar to same-race couples,” says Kao. “What’s different is what they do—or don’t do—in public, like meeting parents or holding hands. Even though they’re breaking the norm, it’s still hard for people to cross social barriers.”

For Kao, an Asian American, the research is a continual reality check. “I tell my students we study misery,” she says. “At a dinner party you can talk with five or 10 friends and come up with a story [about interracial relationships] that works. But the data is not like that, it tells a different story, and it’s kind of depressing.”