Want to change your school’s culture? Start with the right students.

A recent study published in Proceedings of the National Academy of Sciences reveals some important clues on how to change a school’s mores. The gist of the findings? To change school norms, the most well-connected students have to lead the way.

Scholars Elizabeth Paluck, Hana Shepherd and Peter Aronow, who conducted the yearlong project in 56 New Jersey middle schools during the 2012-13 academic year, sought to discover the impact of student-led anti-conflict programs on kids’ behavior. When it was over, they found that groups led by influential students were most successful in changing the way fellow students treated one another. Indeed, in those schools where an average number of  well-connected kids took part in the campaign, reports of student conflict dropped by 30 percent.

These results offer a promising new approach for schools in the midst of a cultural crisis, whether from bullying, cheating or some other undesirable student behavior. To aid educators, the scholars have made their detailed curriculum available online.

The study team designed the investigation with care. To ensure the accuracy of their results, researchers put half of the 56 schools into a control group — which received no specialized anti-conflict programming — and offered the remaining randomly selected 28 schools a carefully designed intervention that sought to reduce friction among the students.

To test the idea that the most socially connected kids have the greatest influence among their peers, the researchers first had to figure out who those kids were. Here, they took a novel approach: All students were given a survey that included every student’s name at that particular middle school. Each child then was tasked with identifying the 10 individuals they’d spent the most time with during the last few weeks, either in person or online. The survey also asked about student perceptions of conflict in school.

This “social network mapping” strategy, which aims to identify the most influential people in a group, differed from typical research methods in two important ways. First, it was driven by students rather than adults; grownups had no role in selecting the influential kids. “When adults pick out students to intervene with, they often pick the popular kids or the traditional leaders,” Paluck said, leaving out some less visible but more influential students, including those who aren’t models of good behavior. Second, by asking kids who they spent time with, rather than who they called friends, the survey revealed which children had the most actual influence.

At each intervention school, randomly selected students were put into a larger group earmarked for anti-conflict training. Some of these students were “social referents,” the social science term that describes those individuals with the most connections to others, and who Paluck calls “the people who you look toward.” For a full year, these groups of students met every other week for specialized training on reducing conflict.

As with the network mapping strategy, the students drove the program. With gentle prompting from a trained research assistant, students in the select groups were asked to speak honestly and openly about problems they perceived in their school. What makes you feel uncomfortable here? What could you or others do about the problem? From these brainstorming sessions, students came up with their own messages to combat the problems they’d identified.

The adult group leader then helped students promote their messages, much like a campaign manager. “We designed activities to help these students make their anti-conflict stance public, noticeable to their peers,” Paluck explained. For example, many groups began wearing wristbands with one-word slogans—“respect” was a popular one—and started hashtag campaigns online. Still others created colorful, highly personalized posters signed by those influential students. The look and feel of these posters bore no resemblance to the typical adult-approved placards that call on kids to be brave nonconformists.

Because every school had its own particular troubles, the messaging differed from one school to another. Even so, some similar themes wove through the students’ anti-conflict communiques: calls for respect, and the demand to “chill out” or “calm down.”

At the end of the year, and after sifting through the volumes of data, researchers discovered their striking results: Those schools that had anti-conflict groups with an average number of well-connected kids saw a 30 percent drop in actual reports of student conflict. Even more notable, schools with the highest number of social referents experienced a 60 percent decline in disciplinary actions. “This is really interesting because we learn that it doesn’t take a lot of the social referent kids to make a big difference,” Shepherd explained.

These findings point to two significant conclusions when it comes to altering school norms. First, to get the most student buy-in on whatever campaign for change a school seeks, the most well-connected kids — as identified by the students — need to take the lead in delivering the message; the better connected the kids, the more likely the campaign will succeed. Second, when trying to figure out how best to convey the campaign message, school leaders must look to these students to create it. The more the solution comes from the kids themselves, in other words, rather than eager adult overseers, the more likely other students will hear and respond to the message. That’s true regardless of the point being delivered, be it against bullying, cheating, using racist or homophobic language, or any other unwelcome teenage behavior that can poison a school’s culture.

Paluck and Shepherd are hopeful that schools will pick up and adapt the curriculum they’ve created to improve their own school environments. “This is a low-overhead program that can be used to address many kinds of issues where student behavior is driven by peers,” Shepherd said.

A trusted guidance counselor or coach could serve as the group’s adult guide, as long as the adult leader “isn’t strongly identified with the administration or with the disciplinary apparatus of the school,” she said. What if school leaders perceive a problem that the students don’t sense? “We advise against ‘telling’ the students what to campaign about,” Paluck said. Still, she added, there’s no reason why a kindhearted adult leader who has demonstrated concern for the kids couldn’t quietly steer the students toward hidden issues affecting their peers.

  • So. They identified “socially connected” eg popular kids and found that by giving them additional attention and opportunities to wield influence, they did. As a result good can be accomplished by steering these popular kids toward resolving unfortunate behaviors among their peers. That sounds great. I suppose no one sees potential for harm in this? No one wonders what happens to those who lead and those assumed to be followers? No one considers the longterm consequences of providing additional adult support and authority and reward to popular adolescents for influencing others?

    Objectively, ultimately, it is the adults who lead. Using older students, barely older students who are nonetheless not adults, has also been shown to be effective effective, but perhaps is less likely to reinforce unequal relationships existing among students of this age. Promoting and exploiting hierarchy because it “works” is short-sighted and unfair. I suspect there is research supporting this. That is, whether the influence is over all “good” or “bad” the favoring of some children over others has enormous capacity for longterm damage.

    Personally, as a child who was not popular at this target age, one who was nevertheless smarter than nearly all the socially-connected kids, I would point out that these middle school/junior high years are miserable enough without adults reinforcing social stratification. Perhaps granting the popular crowd additional authority from adults would not have made my life worse in those years, but somehow I doubt it.

    • A Pickard

      I think
      Maybe you skipped over this entire paragraph in the article.

      This “social network mapping” strategy, which aims to identify the most influential people in a group, differed from typical research methods in two important ways. First, it was driven by students rather than adults; grownups had no role in selecting the influential kids. “When adults pick out students to intervene with, they often pick the popular kids or the traditional leaders,” Paluck said, leaving out some less visible but more influential students, including those who aren’t models of good behavior. Second, by asking kids who they spent time with, rather than who they called friends, the survey revealed which children had the most actual influence.

      • The article summarizes a very large study. They claim it involves over 24000 children, with over 700 “seeds” to influence behavior. The researchers had a worthy goal and the lengthy publication was peer-reviewed. Any successful strategy to improve behavior, specifically to reduce conflict in this case, in middle schools is worth study.

        The seeds were chosen deliberately: “This question is specifically designed to uncover the structure of attention in a
        social network, and identifies social referents as people who are drawing the most attention.” (Paluck et al.)

        This identification IS choosing the most popular kids at this level. Asking students to identify who they recognize as “drawing the most attention” does not alter my concerns about reinforcing inherently unfair social structures at this age. It is a problem at this age. Social structures in middle school, in my experience, are inherently irrational. They are not based on virtue, intelligence, or any other desirable measure, but on conforming to a narrow and arbitrary (and sometimes quite cruel) norm.

        The use of social mapping goes back decades and is often used because the relationships can be damaging. While I applaud the study for re-examing the potential for social structures at this age in hopes of using them to good effect, I still feel as doubtful as encouraged. On the other hand, this was an enormous project, working with many children, and by its own definitions, resulted in high statistical success. I am inclined to play devil’s advocate. My alarm may be without merit. But yes, I did read the entire article, and its source. Did you read my essay, posted above? I offer no graphs of student behavior described like the growth of bacteria in a petrie dish, only personal perception.

        Here is a study also trying to exploit social networking mapping, though with 10th graders http://ccl.northwestern.edu/2008/DoesConnectednessMatter.pdf

        And here is a study about attempts to minimize such connections at middle school age http://www.ncbi.nlm.nih.gov/pubmed/26377235

  • Genesis newton

    its a good idea and technological approach. it has lots of benefits for our backbones kids. everything builds up the quality product.

Author

Linda Flanagan

Linda Flanagan is a freelance writer, researcher, and editor. Her work has appeared in The Atlantic, The Wall St. Journal, Newsweek, Running Times, and Mind/Shift, and she blogs regularly for the Huffington Post. Linda writes about education, culture, athletics, youth sports, mental health, politics, college admissions, and other curiosities. She also reviews books and conducts interviews.

Sponsored by

Become a KQED sponsor