What the Research Says About the Academic Power of Friendship
Gest, who is chair of human services at the Curry School of Education and Human Development, says, “There is a long tradition of informal guidance on how to think about group dynamics in the classroom, but relatively little empirical research to back up particular strategies.” That said, some things are known. There are four big impediments to friendship formation in school: lack of contact, competition, unequal status, and surface-level homophily (a.k.a., “birds of a feather flock together”). Each of these factors can prevent relationships from blossoming, particularly across gender, racial, and other divides. For each of the four roadblocks, teachers wield at least one not-so-secret weapon.
But before getting to solutions, says Barbara Stengel, a professor emerita at Vanderbilt University, who focuses on the philosophy of education, it’s important to think about what friendship really means in a classroom. Aristotle divided the concept into three categories: friendships of utility based on mutual benefit, friendships of pleasure that usually center around a shared interest, and friendships of virtue, the kind with deeper, longer lasting mutual appreciation. When we think of a friend, most of us picture that last sort, the one we can confide in and count on, but the other two types can also make children feel “seen and encouraged,” Stengel says, producing many of the desired academic benefits.
Lack of contact obviously inhibits friendship formation. On the flip side, physical proximity can reduce negative perceptions of a peer. Teachers and administrators often don’t have control over the biggest piece of this puzzle—the makeup of their student body—but they can manipulate contact between the kids they do have. For starters, Juvonen says, teachers and administrators should consider keeping friends together when assigning classes. Schoolwide “house” programs that produce stable cohorts have also shown potential.
Within classes, seating arrangements most directly impact proximity. When children who did not like each other were seated close together for several weeks in one study, their likeability ratings increased. Perhaps they formed Aristotle’s friendships of pleasure, because they were made aware of common interests (comic books!) or maybe the students formed friendships of utility, since whisperings and wisecracks require a set of ears.
Students who dislike one another should not, however, be paired for peer-assisted learning. Most commonly in pairs, peer-assisted learning has been shown to improve the standing of students with learning disabilities and help shy children befriend peers. In choosing dyads, professors Lynn and Douglas Fuchs suggest different strategies for reading and math, both of which involve splitting the class into a top half and a bottom half by current skill level and then choosing one student from each block. But Juvonen says teachers would do well to make these matches with pre-existing friendships and common interests in mind as well, and at least one study backs her up (there, how much partners liked each other predicted how well they learned).
Peer-assisted learning does not, unfortunately, seem to be “sufficient to improve the social integration of children who have behavior issues or whose negative reputation is deeply entrenched,” says Éric Dion, a professor at the University of Quebec at Montreal.
Fostering cooperative learning
Another type of grouping shows promise for that though. By doing away with competition, cooperative learning boosts learning and decreases problematic behaviors, says Cary Roseth, chair of the Department of Counseling, Educational Psychology and Special Education at Michigan State University. It requires establishing positive interdependence, meaning “individuals can attain their goals if (and only if) others in their group also reach their goals,” Roseth has written.
Teachers may require a single finished product from a group (goal interdependence) or may offer a reward to the group if everyone achieves above a certain threshold (reward interdependence). Members of the group can be issued different materials that the group must share to complete the lesson (resource interdependence), or each member of the group could be assigned a different role to play (role interdependence). The group may have its own name (identity interdependence), or each group member may have to complete a different step in a task, like on an assembly line (task interdependence).
When teachers carefully create and scaffold small groups, an expectation that a group member will cooperate arises, and that produces liking. If one group member perceives another as attempting to promote their success, that also promotes liking, even if they ultimately fail. A positive feedback loop results: “The more students work cooperatively to learn, the more they will tend to like each other, and the more they like each other, the harder they will work to help each other learn,” Roseth and colleagues report. In other words, positive interdependence fosters, at the very least, Aristotelian friendships of utility.
Encouraging contact provides the opportunity for friendships to form, but budding connections can easily be nipped by social status asymmetry. Those who don’t conform with school norms on behavior, ability, sexuality, and even body size will be shunned without intervention, Juvonen says. Promoting a cooperative, rather than competitive, learning environment is one step toward redefining “smart” and “good” in children’s minds, but teachers can further decrease status gaps by drawing attention to hidden strengths.
In a 2013 study, when camp counselors encouraged peers to interact inclusively with children who exhibit ADHD symptoms and drew attention to those students’ positive characteristics, the reputations of the children with ADHD improved, and they had more reciprocated friendships. The study’s primary author, Amori Yee Mikami, a psychology professor at the University of British Columbia, stresses that these findings may not translate to the classroom but other studies have shown that teachers voicing a favorable opinion of students and interacting with them warmly tends to increase their social integration.
To this end, teachers should think of themselves in social media parlance as “influencers” or “thought leaders.” Teachers’ relationships with kids “have a big influence on how those kids are seen,” Gest confirms: “Kids who perceive their classmates as not getting along with the teacher come to see those classmates less positively.” But “if teachers make public comments about a child’s academic or social strengths, those have an impact on how kids view that classmate” too.
There’s a problem though: Teachers’ take on who is high status and who isn’t doesn’t always align with kids’, Gest says. “There are kids whom teachers perceive to be disruptive and a problem yet who are quite popular with their classmates. And then conversely, sometimes kids teachers perceive as super nice and prosocial are not particularly influential.” A first step, then, in realizing children’s potential to elevate and inspire one another, is “developing an accurate understanding of what those relationship patterns are.”
One pattern is called homophily. Plato once wrote “similarity begets friendship,” and modern social science research has proven him right. Like tends to stick with like in terms of attitudes and beliefs, but also ethnicity, socio-economic status, and gender even in an integrated classroom. (In Friendship, Denworth reports: “Friendship with opposite-sex peers ‘drops off precipitously after seven years of age.’”)
Yet friendships that bridge these divides have been associated with higher academic outcomes, and Juvonen says, “students with a greater proportion of cross-ethnic friendships reported lower vulnerability” to peer victimization. On the other hand, discriminatory experiences lead to anger, impulsivity, depression, anxiety, sleep loss, and more, all conditions that drive down academic engagement and performance.
For cross-group friendships to thrive, Juvonen says, teachers and administrators have to “disrupt typical social dynamics and avoid instructional practices that highlight differences.” Going after low-hanging fruit, Juvonen recommends we stop saying, “Good morning, boys and girls.” Using these categories implies that they have functional importance in elementary school (when research has yet to prove they do) and impedes same-gender bonds.
Administrators can also consider explicit anti-bias interventions. Juvonen says a puppet program that “teaches about acceptance of various body shapes has been shown effective in reducing negative attitudes and stereotypes about larger body shapes.” Inclusive curricula can also alter social dynamics.
Though initiatives like these take time and institutional support, there’s one thing educators can do right away, Laursen says. While perceived similarities predict who will become friends better than actual similarities, it’s the latter that determines whether friendships will last. Teachers can help kids’ friendship calculus be more accurate by making less obvious similarities salient. Another way of looking at it? By drawing attention to traits and interests that aren’t as readily apparent as gender or skin tone (e.g., “You two and your Minecraft obsession!”), teachers foster Aristotelian friendships of virtue.
Juvonen says extracurricular activities like sports and interscholastic robotics competitions provide the ideal context both for highlighting shared interests and promoting positive interdependence, but access is often a problem. Administrators can try to decrease hurdles such as transportation and out-of-pocket expenses, as well as ensuring there’s extra support on hand to facilitate the participation of special needs students. But logistical stumbling blocks aren’t the only type. “Some kids are just reluctant to take the big step to join a club,” Laursen says, and schools would do well to create an emotionally safe environment. That can mean paired activities and inclusion-oriented clubs such as Gay-Straight Alliances.
Kids can also be encouraged to find hidden similarities on their own. Julia Smith, who teaches first-grade in San Francisco, reads her students The Day You Begin by Jacqueline Woodson:
There will be times when you walk into a room and no one there is quite like you until the day you begin to share your stories. My name is Angelina and I spent my whole summer with my little sister, you tell the class …. Your name is like my sister’s, Rigoberto says. Her name is Angelina, too…. This is the day you begin to find … every new friend has something a little like you.
But Elizabeth Self, an assistant professor at Vanderbilt University, says it’s important to keep in mind there simply isn’t enough research on encouraging cross-group friendships for academics like her to provide a 10-tricks book. For the most part, they are instead “going to talk about, you could do this, but you’d need to watch out for that.”
Case in point: Just how much to spread kids out.
Skill sorting and ability grouping, Juvonen says, “not only reduces contact, but also highlights status differences between demographic groups.” Tracked classes, resource rooms, and second-language learner programs that separate groups of students and highlight their differences are also “likely to hinder peer acceptance and the development of friendships,” she says.
And yet, distributing a small group of atypical kids across classrooms can also be the wrong call. In one study, children with disabilities, who can struggle with social integration, were just as likely to have friends and be accepted as their developmentally typical peers when placed in classrooms where one-third of the students had a mild disability. Juvonen’s conclusion: “There is a critical minimum mass required for groups of vulnerable students to be socially integrated.”
Research on race relations in middle and high schools suggests exactly that. In Why Are All the Black Kids Sitting Together in the Cafeteria? Beverly Daniel Tatum, a psychology professor and former president of Spelman College, explains that around the onset of puberty, Black students start to explore their identity just as “the world begins to reflect their Blackness back to them more clearly.” In racially mixed settings, she writes, voluntary “racial grouping is a developmental process in response to an environmental stressor, racism.” When it comes to racial microaggressions, white peers “are unprepared to respond in supportive ways.” That makes joining with other Black students “a positive coping strategy.”
A teacher with a class of 25 students that includes 5 Black students and needs to be split into 5 groups may be tempted to create diverse pods by placing one of the underrepresented students in each group, but doing so can actually set intergroup relations back. Once kids are old enough to grapple with race, numerical insignificance and stereotype threat—which one of Dr. Tatum’s young sources described as “that constant burden of you always having to strive to do your best and show that you can do just as much as everybody else”—can silence and alienate Black children, reduce their status, and thwart friendship formation. When small groups involve peer critique, preventing critical mass can also leave Black students emotionally unprepared to receive feedback. As counterintuitive as it may seem, allowing Black students “the psychological safety of their own group” can actually increase the likelihood that they form friendships outside it.
Elizabeth Self says similar concerns apply to “putting kids from the same linguistic background together in maths small group work.”
Making game-time calls
At the end of the day, teachers will have to make judgment calls when it comes to friendship. Students who are easily distracted may benefit from more individual work, and there’s research showing that friends do interfere with productivity in some circumstances: for example, when they’re not engaged by the subject matter or they put one another’s feelings over giving meaningful feedback. But if a friendless child goofs off with a peer, Laursen says, a little more leeway may be in order, since research shows that kids with at least one friend are both less likely to be bullied and less harmed by bullying. It would make sense then, to seat a child with very low social status near one who is both friendly and popular. A warm relationship with someone like that could increase classwide acceptance considerably.
Elizabeth Self likes the idea of reconceptualizing friends as a resource, thinking, “How can we give them permission to draw on that person?” When a student is getting out of sorts, for example: “If they have a good bud who is not in the classroom, say: ‘Let’s go see if we can pull Margarita from Ms. Jon’s class. You all stay in the hall for five minutes. We are going to set a timer to see if spending some time together helps you to be able to come back into class.’” In the context of restorative justice circles, why not have an ally present for each child? “I think there is rich opportunity here,” she says.
But Gest wants to remind teachers, administrators, and their communities: “You can’t address everything at once, through either a seating arrangement or a group learning assignment.” Yes, friendship can present untapped academic potential, but “there’s limits to how much teachers can do.”
This article is part of the “Friendship in Schools” series, which explores the complexities of friendship at various stages of learning.