Stanford researcher explores why cliques thrive in some high schools more than others

Educators may be able to affect teen cliquishness through changes in school organization. Schools that offer students more choice are more likely to be rank-ordered, cliquish and segregated by race, age, gender and social status.

iStock High school girls talking

Daniel A. McFarland, a professor at Stanford Graduate School of Education, studies the way students self-segregate at high schools.

At almost any American high school, the elements of teen social networks are readily apparent: the cliques, the pecking orders and the varying degrees of self-segregation by race, age, gender and social status.

For years, sociologists have identified seemingly universal human instincts that spur this kind of sorting. These include the desires for familiarity and certainty, for control and dominance, and for security and support.

But as ubiquitous as those instincts are, students in some schools form more cliquish, hierarchical and segregated social structures than in others. What accounts for the variation?

It turns out that the organizational setting of a school itself, its "network ecology," has a big impact. Schools that offer students more choice – more elective courses, more ways to complete requirements, a bigger range of potential friends, more freedom to select seats in a classroom – are more likely to be rank-ordered, cliquish and segregated by race, age, gender and social status.

By contrast, pecking orders, cliques and self-segregation are less prevalent in schools and classrooms that limit social choices and prescribe formats of interaction. Smaller schools inherently offer a smaller choice of potential friends, so the "cost" of excluding people from a social group is higher. In addition, structured classrooms guide student interactions in prescribed routes and encourage students to interact on the basis of schoolwork rather than on the basis of their external social lives.

Those are among the conclusions of a new study, published online today in American Sociological Review. The lead author is Daniel A. McFarland, professor of education at Stanford Graduate School of Education.

"Educators often suspect that the social world of adolescents is beyond their reach and out of their control, but that's not really so,'' McFarland said. "They have leverage, because the schools are indirectly shaping conditions in these societies."

Datasets on friendships

The study draws on an analysis of two datasets about friendships. One considers friendships at the classroom level and the other at the school level. At the classroom level, the researchers tapped into detailed data of friendships and social interactions compiled by McFarland at two very different high schools over a two-semester period. The school-level data came from the National Longitudinal Study of Adolescent Health.

The researchers found that large schools tend to accentuate the quest by adolescents for friends who are similar to themselves, an instinct that sociologists call "homophily." Bigger schools offer a broader range of potential friends, as well as greater exposure to people who are different. It's a mixture of freedom and uncertainty that spurs students to cluster by race, gender, age and socioeconomic status.

But a school's size is only one factor. The researchers also found that a school's openness to choice spurs cliques and social-status hierarchies as well.

In schools with a strong focus on academics, where teachers have a hand in setting the pace and controlling classroom interactions, teenagers are less likely to form friendships based on social attitudes imported from outside the school. Instead, friendships are more likely to develop out of shared school activities and similar intellectual interests.

As the researchers put it, a positive educational climate strengthens the school's "system membrane" and makes it more impervious to external criteria for friendship such as race or social status. In other words, a more rigid school setting can sometimes promote more open-mindedness in making friends — a potentially valuable quality in adulthood.

McFarland cautioned that the study doesn't mean that students are necessarily better off in small schools with less choice. For one thing, the practice of putting students on particular tracks based on their apparent academic prowess often has the side effect of segregating students according to race. A bigger and more diverse student population may well foster self-segregation, but a smaller and more elite school is almost inherently more segregated in the first place.

Beyond that, the likely influence of these structural topographies may be complicated and contradictory. Different kinds of students are likely to thrive in settings with different blends of supervision, freedom and uncertainty.

"We're not proposing that we all go to a forced boarding-school model,'' he said. "The truth is that we are not sure which kind of adolescent society is best for youth social development, let alone what position in them is best." 

More studies on the way

The main goal of this study, he continued, is to shed light on how a school's environment affects the shape of adolescent social networks. The next round of studies, he said, will look at which kinds of social networks and social networking positions best help adolescents prepare for adulthood.

"There likely isn't a simple answer," McFarland said. "What may work well for a shy child may not work well for a gregarious one, and neither solution may prepare them well for the realities of adulthood. We just need to study it and see."

The study, Network Ecology and Adolescent Social Structure, was supported by the National Institutes of Health, the National Science Foundation, Duke Network Analysis Center and Stanford's Center for Computational Social Science. McFarland's co-authors are James Moody of Duke University and King Abdulaziz University, David Diehl of Vanderbilt University, Jeffrey A. Smith of the University of Nebraska-Lincoln and Reuben J. Thomas of the University of New Mexico.

Dan McFarland, Graduate School of Education:

Jonathan Rabinovitz, communications, Graduate School of Education: (650) 724-9440,

Dan Fowler, media relations, American Sociological Association: (202) 527-7885,