MAZUR AND TREISMAN offer striking stories of transformation. The structure of the classroom has been transformed, as have student interactions and individual achievement. It is impressive to learn that all students improved their test scores after Mazur and Treisman introduced collaborative-learning models. But it is perhaps more striking still to view these transformations together and realize that Mazur and Treisman have created something larger: a classroom culture shift. This culture shift affects students both internally, in how they define their values, and externally, in how they interact with peers.
Internally, Mazur and Treisman encourage students to redefine their personal understandings of intelligence and merit. They reject the dominant narrative about intelligence: that it is fixed, inherent, and measured by tests. Instead, Mazur and Treisman teach students to view intelligence as malleable and based on success in collaborative work. “Intelligence” is more expansive than a raw test score, and “merit” encompasses positive character traits such as communication, collaboration, and group leadership. Mazur and Treisman deliberately design lectures and classroom activities to persuade students about two key ideas: that intelligence involves hard work, yes, and that intelligence also involves interpersonal skills. Intelligence, in a democratic and meritocratic context, is never static.
The second shift in Mazur’s and Treisman’s classrooms involves external relationships. These professors encourage students to redefine their relationships with each other. Instead of working alone, students are encouraged to work together. Students become teachers and teammates. Classes show students that diverse perspectives can collectively lead to better solutions. Competition is replaced with collaboration: rather than focusing on their own isolated intelligence, students began to value and seek out new peer perspectives and partnerships.
These internal and external transformations shift the classroom culture toward a celebration of democratic merit. But observing this change is just the starting point. My question is how. How did Mazur’s and Treisman’s practices produce such results?
The world of social science helps us answer the question of what leads to success. Here I want to present six thinkers, six different researchers who approach this question from different angles. Carol Dweck and Paul Tough take up the question on an individual level by asking, What mind-sets, behaviors, and attitudes will lead to one individual’s success and another’s failure? The research of Anita Woolley and Scott Page addresses the question at a group level: How can a particular group achieve higher results than either other groups or individuals working on their own? Marilynn Brewer has looked at how the intersection between the individual and the group affects success. And, finally, Jo Boaler has explored an example of these successful approaches in the math classrooms of an urban public high school. Each of these thinkers casts light on a different dimension of how a culture shift can occur in classrooms like Mazur’s and Treisman’s and, perhaps most important, how we can spread these culture shifts to new people and places.
HOW DID MAZUR and Treisman change student test scores? Research suggests that test scores improved because a fundamental change was at work: a shift in mind-set. Both Mazur and Treisman decided to change how they spoke about “answers”: they encouraged students to value the learning process, not just the final score. By changing the values in the classroom, Mazur and Treisman worked to redefine the measures of merit.
Both Mazur and Treisman taught students that their aim was not to get the “right answer.” Instead, the goal was to understand how to get to the right answer. Students needed to explain their approach to a problem and articulate any obstacles they encountered. These professors made it clear that students should not be embarrassed if they gave a wrong answer or had no answer at all. Students were urged to view giving wrong answers as an opportunity for higher understanding. They needed to embrace their struggle so that they could understand their missteps and learn better, alternative approaches. In this way, Mazur and Treisman rejected the concept that merit is defined by a final test score. Instead, they each rebuilt the platform of a meritocracy, using Amartya Sen’s concept of merit. They redefined merit as an opportunity for both continual self-improvement and self-reflection, as well as the development of collaborative approaches that stimulate learning.
After creating a system that values learning processes over right answers, both Mazur and Treisman witnessed a rise in their students’ test scores. Why would a pedagogy that values self-improvement over test scores yield an increase in student performance?
Dweck and Tough may have the answer. In their work, these researchers set forth to answer the puzzling question: Why do some children succeed while others lose their way? Dweck, a psychology professor at Stanford University, has approached this problem by designing tests to measure patterns of student motivation. Tough, a writer interested in educational reform, has approached the puzzling question by looking at methods that enable low-income children to reach higher levels of success.
Though their methods vary, Dweck and Tough have reached similar conclusions. They both found that children will be more successful if they believe their intelligence is capable of growth, rather than believing intelligence is innate. Tough has added to Dweck’s observations by concluding that a child will also be more successful if she or he focuses on building character strengths in a similar way, viewing them as malleable rather than fixed. Understanding Dweck’s and Tough’s conclusions allow for a better understanding of why Mazur and Treisman were so effective when they changed the curricular culture in the classroom.
CAROL DWECK’S NUMEROUS studies show that an individual who believes intelligence is “fixed” is much more likely to fail in the face of new challenges, while an individual who believes that intelligence can grow with hard work is much more likely to excel in the face of new challenges.
Dweck first investigated the underpinnings of human motivation as a graduate student at Yale University in the 1960s.1 She had read about “learned helplessness” in animals. Psychologists at the University of Pennsylvania showed that after repeated failures to stop something negative from happening, most animals conclude that the situation is hopeless and beyond their control. After such an experience, the animal often remains passive even when it can effect change, a state the researchers called “learned helplessness.”2
After reading these studies, Dweck observed that some people also exhibited “helplessness” in the face of repeated failure. But she was more fascinated by the people who don’t—the people who still persevere in the face of setbacks. She wondered: “Why do some students give up when they encounter difficulty, whereas others who are no more skilled continue to strive and learn?”3 After more than three decades of conducting countless studies, she concludes that “one answer. . . [lies] in people’s beliefs about why they had failed.”4 In short, students who attribute poor performance to a lack of innate ability will continue to perform poorly. However, students with the same incoming abilities who believe their poor performance is due to effort—and thus can be overcome with hard work—will improve their performance over time.5
Dweck has developed a broader theory of what separates the two general classes of learners: helpless versus mastery oriented. Through her research, she found that students not only diverge in how they explain their failures but that they also hold different “theories” of intelligence. Some subscribe to a “fixed mind-set” of intelligence, others to a “growth mind-set.”
Dweck coined the term “fixed mind-set” to describe “the helpless ones [who] believe that intelligence is a fixed trait: you only have a certain amount, and that’s that. . . . Mistakes crack their self-confidence because they attribute errors to lack of ability, which they feel powerless to change.”6 Students with a “fixed mind-set” avoid challenges because they want to avoid mistakes.7 They spend their time documenting their intelligence or talents instead of developing them.8 In contrast, students with a “growth mind-set” believe that intelligence can be developed through education and hard work. “They want to learn above all else,” Dweck writes. “[Since] slipups stem from a lack of effort, not ability, they can be remedied by more effort. Challenges are energizing rather than intimidating; they offer opportunities to learn.”9 To illustrate the behavior of “growth mind-set” students, Dweck shared several anecdotes from her studies:
[Growth-minded students] focused on fixing errors and honing their skills. One advised himself: “I should slow down and try to figure this out.” Two schoolchildren were particularly inspiring. One, in the wake of difficulty, pulled up his chair, rubbed his hands together, smacked his lips and said, “I love a challenge!” The other, also confronting the hard problems, looked up at the experimenter and approvingly declared, “I was hoping this would be informative!”
Dweck wanted to know whether a growth mind-set led to higher, lower, or equal levels of student performance. To answer this question, Dweck did what she was trained to do: she designed an experimental study, and then another, and then another. All the results repeatedly reaffirmed her conclusion that students with a growth mind-set have greater academic success and are more likely to outperform their fixed-mind-set counterparts. For example, in 2007, Dweck and her colleagues shared the results of a two-year study that monitored 373 students during the transition from elementary to junior high school, when classwork gets more difficult. At the beginning of seventh grade, all the students had roughly the same incoming math scores. But though students had equivalent scores, they did not have equivalent mind-sets—some saw intelligence as fixed, while others believed it could “grow” with effort. To assess these students’ initial mind-sets, Dweck and her colleagues asked students at the start of seventh grade to agree or disagree with statements such as “Your intelligence is something very basic about you that you can’t really change.” After determining each student’s initial mind-set, Dweck and her colleagues tracked each student’s grades. The results? Not only did students with “growth mind-sets” perform better on math tests than their fixed-mind-set peers, but this difference in performance also grew over time.10
Furthermore, Dweck’s research finds that even our “mind-set” is not innate but malleable. The type of praise an individual receives can affect an individual’s mind-set. When students receive praise for their intelligence, they are more likely to adopt a fixed mind-set than when they receive pats on the back for effort. In a 1998 study, Dweck found that those who were congratulated for their intelligence
shied away from a challenging assignment. . . far more than the kids applauded for their effort. . . . When we gave everyone hard problems anyway, those praised for being smart became discouraged, doubting their ability. And their scores, even on the easier problem set we gave them afterward, declined as compared with their previous results on equivalent problems. In contrast, students praised for their effort did not lose confidence when faced with the harder questions, and their performance improved markedly on the easier problems that followed.11
Dweck’s studies show us two things. First, students who view intelligence as capable of growth will perform better than those who don’t—that is, valuing the learning process over a test score will help students achieve at higher levels. Second, students are more likely to adopt a growth mind-set if others praise them for their effort rather than their intelligence. These conclusions mirror the results of Mazur and Treisman, who in effect increased student test scores by adopting strategies that promoted a growth mind-set over a fixed mind-set in their classrooms. As if they had been reading Dweck’s work, both Mazur and Treisman placed great value on the learning process, not just on giving a right answer.
BUT THIS IS not the end of the story. Mazur and Treisman did more than just convey to students that intelligence was malleable rather than fixed. They also conveyed that success depended on certain study habits and behaviors that were—like intelligence—capable of improvement.
This focus on academic behaviors, or “character traits,” may be another secret to student success, according to author Paul Tough. Like Dweck, Tough is interested in what makes some students succeed while others flounder. He decided to write a book that would help him “solve some of the most pervasive mysteries of life: who succeeds and who fails? Why do some children thrive while others lose their way? And what can any of us do to steer an individual child—or a whole generation of children—away from failure and toward success?”
His research question was similar to Dweck’s. But while Dweck turned to social science for the answer, Tough decided to take a more holistic approach by showing the connections between very different fields of research: economics, neuroscience, pediatrics, and psychology. Surveying these various fields, Tough made a bold observation: the prevailing “cognitive hypothesis” that saturates our culture is misguided.12 This hypothesis, he explains, is “the belief, rarely expressed aloud but commonly held nonetheless, that success today depends primarily on cognitive skills—the kind of intelligence that gets measured on IQ tests, including the abilities to recognize letters and words, to calculate, to detect patterns.”13
This hypothesis is exactly what gives the SAT and other standardized tests their life’s breath. Being able to half-decipher the meaning of an arcane word or to eliminate one or two wrong math answers and thereby improve one’s ability to guess—without knowing how one has arrived at the right answer—these are the skills that our culture prizes. Yet the research and testimony that Tough encountered thoroughly discredit this “cognitive hypothesis.” In its place, he has developed a new theory based on the importance of positive character traits. Tough found that what matters most to a child’s success is not how much information we can stuff into a child’s brain but rather “whether we are able to help her develop a very different set of qualities, a list that includes persistence, self-control, curiosity, conscientiousness, grit, and self-confidence.”14 Thus, Tough appropriately titled his book How Children Succeed: Grit, Curiosity, and the Power of Hidden Character.
Tough offers many studies and programs to support his assertion that success depends upon developing key character traits more than on simply developing rote skills. To show that IQ scores don’t summarize intelligence, he gives the example of the Perry Preschool project, conducted in the 1960s. In this research project, children were recruited from a low-income neighborhood. Half were admitted to Perry, a high-quality, two-year preschool program, while the other half served as the “control group” without a free offer of preschool. The researchers followed the two groups of children for decades to track the effects of the preschool program. The results were not what you might expect. Researchers found that attending Perry Preschool had no lasting effect on the children’s IQ scores. By third grade, both the Perry children and the control group had equivalent IQ scores. Researchers did observe, however, that attending Perry Preschool carried an important long-term effect. Compared to the control group, Perry students were more likely to have graduated from high school, more likely to be employed by age twenty-seven, more likely to be earning more than $25,000 a year by age forty, less likely ever to have been arrested, and less likely to have spent time on welfare.15
These results intrigued James Heckman, a professor of economics, who dug deeper into the archives. Heckman found that the Perry Preschool scored students on what Heckman called “noncognitive skills,” such as curiosity, relationships with fellow students, social fluidity, and self-control. Heckman found that these noncognitive factors were responsible for as much as two-thirds of the total benefit that Perry gave its students and led, from his perspective, to the differences in measurable life outcomes.
From the Perry Preschool study, Tough concluded that social skills “and the underlying traits they reflected turned out to be very valuable indeed.”16 Tough has been able to show that character traits are important not only in an early childhood environment but in a college context as well. In his chapter “How to Build Character,” he writes about David Levin, a founder of KIPP Academy. KIPP (which stands for the Knowledge Is Power Program) is an innovative charter school program whose mission is to provide a high-quality middle school education to children from low-income communities. Since its start in the 1990s, KIPP schools have yielded some highly impressive student scores. In 1999, the students of KIPP Academy in Bronx, New York, earned the highest scores of any school in the Bronx and the fifth-highest in all of New York City.
But Levin came to the same conclusion I am advocating in this book: test scores in and of themselves, as a single-minded approach to education, are meaningless. The initial cohort of KIPP in New York City did extremely well in terms of their test scores. They left middle school with outstanding academic results and most won admission to highly selective private or Catholic high schools. Almost every member of the Bronx class made it through high school, and most enrolled in college. But once in college KIPP students started to struggle: six years after their high school graduation, only 21 percent of the cohort had completed a four-year college degree.17
Levin was pained by this low graduation rate. He analyzed the dropout reports and noticed something curious: the students who succeeded in college were not necessarily those who had tested well or excelled academically in KIPP. Instead, they seemed to be the ones who possessed certain other gifts, skills like optimism and resilience and social agility. Levin knew then that he had to promote these traits among his students.
Levin worked closely with social psychologists Martin Seligman and Angela Duckworth to identify key character traits that would enable students to reach the highest levels of success, both in KIPP and later in college. After much discussion, they identified seven traits: grit, self-control, zest, social intelligence, gratitude, optimism, and curiosity. After identifying the traits, Duckworth worked with Levin to develop an evaluation tool that teachers and students could use to assess “character strengths.” Then, in 2007, Levin introduced the first-ever “character report card.” He knew there would be problems with quantifying character, so he asked students to rate themselves, as well as having teachers rate them. By 2011, Tough observed that the emphasis on character traits had gone “viral”—it was everywhere at the KIPP Academy in New York. Levin explains that the focus on character “has to permeate everything in the school, from the language people use, to lesson plans, to how people are rewarded and recognized, to signs on the wall. If it’s not woven into the DNA of an institution, it will have minimal impact.” Indeed, KIPP’s increased emphasis on character has shown promising results. Though the first cohort of graduates was in 2003, all the academies, across the United States, saw only a 21 percent college graduation rate in six years. However, the class of 2005 doubled this graduation rate: 46 percent of KIPP students graduated in six years.18
The effectiveness of KIPP’s character-building approach will become more apparent with time, but this example highlights the importance of developing particular traits in order to excel in college and beyond: resilience, optimism, and social agility. These traits are important not only for low-income students but for all students; they are essential ingredients for success. But the prevailing focus on test scores downplays the importance of these character traits—and it is the students who suffer the most because of it. Emphasizing an individual test score discourages students from taking risks, engaging in creative solutions, and embracing failure as an opportunity for learning.
Dominic Randolph, the headmaster of the prestigious Riverdale Country School, a prep school in New York City, shares the same concerns about the importance of teaching students character traits such as resilience and curiosity. Despite the different demographics of KIPP Academy in the Bronx and Riverdale (whose students primarily come from wealthy and highly educated families), Randolph worries that Riverdale students are missing out on strengthening their character, by which he means developing traits like grit and self-control that will carry a student through struggles and failure. He criticizes the over-emphasis on IQ testing, observing that “this push on tests. . . is missing out on some serious parts of what it means to be a successful human.”19 According to Randolph, a culture that focuses primarily on test scores fails to develop these critical character traits in its young people. It does not encourage students to take risks, develop curiosity, or engage in activities that may lead to failure and thereby growth. He explains, “People who have an easy time of things, who get 800s on their SAT’s, I worry that those people get feedback that everything they’re doing is great. And I think as a result, we are actually setting them up for long-term failure. When that person suddenly has to face up to a difficult moment, then I think they’re screwed, to be honest. I don’t think they’ve grown the capacities to be able to handle that.”20
Unfortunately, Randolph’s worries don’t seem that far-fetched. I have already pointed out that the job choices of many graduates from our most elite colleges show a lack of empathy and creativity. In fact, the negative social repercussions of raising a generation of graduates afraid of bold pursuits that require some degree of struggle and potential failure could in the long run be devastating. In Tough’s closing chapter, he observes: “There are fewer entrepreneurs graduating from our best colleges these days: fewer iconoclasts, fewer artists, fewer everything, in fact, except investment bankers and management consultants. Recently, the New York Times reported that. . . more than half of the [Princeton] class was going into investment banking or consulting—and this after the near-collapse of the finance industry in 2008.”21 James Kwak, an economics blogger and law professor, explained this trend toward banking and consulting among graduates of elite schools: “It’s that the firms make the path and the decision so easy to take and so hard to resist.”22 Kwak concludes, “For people who don’t know how to get a job in the open economy and who have ended each phase of their lives by taking a test to do the most prestigious thing possible in the next phase, all of this comes naturally.”23 A decline in creative leadership and problem solving is one possible side effect of emphasizing test scores over character traits like resilience, social intelligence, and curiosity.
The twenty-first century needs college graduates who can address the prevailing issues of our era, such as global warming, an expanding technological landscape, and the equitable distribution of opportunities. These issues require collaboration, experimentation, creativity, and optimistic perseverance; this kind of thinking will help us tackle old problems in ways that will provide us with new solutions.
MAZUR AND TREISMAN encouraged their students to value the learning process over a right answer. They emphasized the importance of openly discussing failures to move toward solutions, creatively approaching a problem by experimenting with different approaches, and recognizing that struggle was the only path to improvement. These principles fostered within students character traits such as resilience, optimism, and grit that often led to success. The development of these traits may partially explain why the students in Mazur’s and Treisman’s classrooms were far more successful at the end of the semester. It may also explain why Treisman’s students were more likely to concentrate in mathematics as their college major and more likely to participate in, and lead, activities throughout the campus community.
But the transformations in our students are not only internal. We can also effect an external transformation that encourages student collaboration and interdependent learning—the roots of democratic merit. Research sheds light on how student collaboration can substantially improve overall performance when students teach each other, learn from one another, and value differing perspectives. However, the research also shows that performance does not automatically improve if people are put in groups; group dynamics matter. Our next four thinkers: Anita Woolley, Scott Page, Marilynn Brewer, and Jo Boaler help explain the dynamics that enabled the collaborations in Mazur’s and Treisman’s classrooms to be so effective.
ANITA WOOLLEY STUDIES INTELLIGENCE. She’s a professor at Carnegie Mellon University in the department of cognitive science, a discipline that studies how the mind works. Woolley, however, is unlike most of her colleagues. Instead of studying the brains of individuals, she studies the “brains” of groups. Intelligence, for Woolley, is measured by the ability of groups. She calls this “collective intelligence.”
In today’s world, it is increasingly common for groups, rather than individuals on their own, to solve complex problems. A huge part of the reason for this is the complexity of the problems themselves. The world is getting more and more multidimensional, and it is virtually impossible for any one person to understand all the sides of a problem. People need to work together. Businesses organize themselves into teams to come up with marketing strategies. Government officials meet to devise national security plans. Scientists share information across large networks to develop new vaccines. In today’s world, almost everyone is collaborating.
In a collaborative society like ours, measuring intelligence only in individual people—as cognitive scientists have traditionally done—seems to miss the point. Woolley has believed this from the start of her career. When everyone’s working together, we should be concerned about how and why groups become smart. It makes little sense to focus so much on the intelligence level of the individual when most knowledge building and problem solving today happens within teams. Woolley wanted to delve deeper into what accounts for collective intelligence.
In beginning this endeavor, Woolley first asked, are groups smartest when they are made up of the smartest members? So, if we wanted to put together the brightest group, the one that would be the most effective at solving the problems of, say, government, or business, or science, would we simply try to find the smartest individuals and bring them together? This is a logical and traditional belief about group intelligence: throw the smartest people you can find into a room, and they’re bound to be smart together.
Woolley set out to determine whether this widely held belief was true. She put together different combinations of individuals with varying IQ levels, genders, backgrounds, and so forth, and asked them to perform a variety of tasks: some simple, some hard. The groups worked away. In observing them as they worked, Woolley noticed that the groups acted very differently from one another. In some groups, everyone spoke and was encouraged to share ideas. In others, one or two people dominated the conversation. Each came to have a culture of its own.
By the end of the experiment, the performance of the groups varied dramatically. Some were very adept at solving all sorts of problems and tasks; some were not adept at all, and others were somewhere in between. In short, some teams were smarter than others. But why? Was it simply as everyone had believed, that the smartest groups were the ones with the smartest members?
What Woolley found was striking. The cumulative intelligence of the groups’ members—if you were to add up the IQs of everyone in the group—was not predictive at all of the intelligence of the group as a whole. It didn’t seem to matter whether the group was made up of the smartest individuals or of the least smart individuals. What mattered, instead, were a whole host of other factors.
Surprisingly, Woolley found that the sex of the members (largely because of that the group’s internal social dynamics) made a huge difference on the outcomes. Groups with higher percentages of women performed better than groups with higher percentages of men. The “female effect,” Woolley found, can be partly explained by women’s tendency toward greater social sensitivities. Women were more likely to read nonverbal cues and perceive accurately what others were feeling or thinking. They were also more likely to encourage turn-taking in the discussions. The cooperation that women fostered led groups to take advantage of the skills and knowledge of all the members. Those attributes were brought out and shared, and then multiplied, in a way that didn’t happen in groups with more men and less cooperation. Communication and inclusivity were key to the groups’ success—far more so than the overall intelligence of the members.
Woolley’s findings were consistent with other research that’s been done on the role of gender in group processes. In a study of groups working in business settings, Graham Fenwick and Derrick Neal found that groups with equal numbers of men and women, and/or with higher numbers of women than men, performed better than groups with higher proportions of men.24 The reason for this, they found—as did Woolley—was the positive effect women had on the ability of the group to collaborate. The presence of women was shown to encourage and facilitate group processes so that all members felt comfortable participating and sharing their knowledge and skills.
The consequences of Woolley’s findings are huge, with implications for business, government, and the sciences. With so much happening in groups, as it does in today’s world, Woolley’s work suggests that our efforts would be better directed toward building group intelligence. Group or “collective” intelligence is not simply a direct outgrowth of individual intelligence. Rather, it grows and develops in unique ways that have to do with social sensitivities, group norms, and internal team dynamics. Efforts to improve productivity, efficiency, or creativity should involve teams with a high proportion of women and be structured to encourage inclusive, participatory group norms.
For education, the findings are even more significant. They suggest that our heavy focus on building individual abilities—as measured so carefully in our testocratic world—might be misplaced. Rather than focusing excessively on enhancing individual IQs, as today’s educational system does, we need to improve—and reward—students’ abilities to work in groups. This is the challenge we must meet if we want to prepare our next generation to solve the complex problems of today’s and tomorrow’s worlds.
ACROSS THE COUNTRY from Anita Woolley, Scott Page came to a similar realization in his research: the highest-performing group was not composed of the highest-performing individuals. While Woolley uncovered the importance of female participation to group success, Page uncovered the importance of diversity, where “diversity” refers not to what we look like on the outside but to our distinct tools and abilities. In his book The Difference: How the Power of Diversity Creates Better Groups, Firms, Schools, and Societies, Page recounts that he “stumbled upon” the power of diversity.25 This happened during his “first real job,” as an assistant professor of economics at the California Institute of Technology in Pasadena.
The year was 1995. At the time, Page had no intention of building his career around mathematical models of diversity. He was simply enjoying the fact that he could wear shorts to work every day, while exploring Caltech’s abundant resources. Page did have a budding interest in how groups performed as problem solvers. Like other economists, Page had made two observations. First, he saw that our modern world demanded complex problem solving to contend with multidimensional issues like addressing environmental policies, designing social welfare systems, and cracking secret codes. Second, he saw that groups were more adept than individuals at solving these multidimensional problems. However, Page had not yet explored what types of groups were the “best” problem solvers.
One winter evening, Page decided to have what he describes as “a little fun.” He constructed “a computing model of diverse problem solvers confronting a difficult problem.” To create “groups,” Page selected members from a pool of problem solvers who had received different test scores on an assessment. He formed two different types of groups: one of “high performers” and the other a “diverse” group of problem solvers. The “highest performers” group comprised members with the highest test scores. The second, “diverse” group was made up of a random sampling of members with a range of test scores: some high, some low. These varying test scores reflected what Page called different “tools,” or ways of thinking—a fascinating way of looking at democratic merit. Page then created a mathematical formula to test each group’s ability to solve difficult problems. From this model, Page observed that the random sampling of diverse performers outperformed the highest performers every single time.
This may sound counterintuitive. You mean the brightest don’t always do the best? In Page’s model, diversity trumped ability. In truth, Page said, he was startled by his own finding. And he wanted an explanation. So he partnered with his colleague Lu Hong to try to explain this surprising discovery. Unlike Woolley, in her social science study, Page and Hong used a different method of proof: mathematical formulas. They showed that the “highest performers” possessed the same “tools” for, or approaches to, problem solving, whereas individuals who received a range of scores exhibited different tools or approaches. A group would be most successful when its members could share their diverse approaches to a problem.26 Through his study, Page built upon the saying “two heads are better than one” to show that two heads with diverse approaches will outperform two heads with the same approach but higher test scores.
Page points out one of the limits of his finding: diverse groups will only outperform homogeneous groups if the members share their different approaches with one another. A group’s social dynamics are relevant; witness studies such as Woolley’s, which focus on how group norms affect performance. But Page’s research is also important because it demonstrates why our society should promote diversity for reasons besides “fairness.” Today’s world demands problem solvers. Groups outperform individuals in problem solving, and Page shows that diverse groups—with diverse approaches—will yield the best solutions to modern dilemmas.
IN ADDITION TO wanting to be “ourselves” as individuals, humans want to be part of groups. This desire takes on many different forms. We organize into families, communities, countries, nationalities, and ethnicities. As we come to see ourselves as part of these groups, we adopt a number of different “social identities.” Every day, people identify themselves by referencing their membership in all sorts of groups—by gender, ethnicity, college alma mater, hometown, participation in a club or activity. The list goes on. What these different group-based identities reflect is the desire to “belong” or to feel part of a group, and one aspect of group-belonging that people particularly like is how belonging to a certain group makes them different from members of other groups. While being part of a group means that you share things in common with fellow members, it also means that those things make you stand out. For example, sports fans both love their team and their fellow fans and think that their team’s rivals and their fans are despicable. Red Sox fans both love other Red Sox fans and hate Yankee fans.
So what does this have to do with education? Well, as it turns out, how much you perceive your membership in the “in-group”—whether you’re a Red Sox fan living among other fans in Boston or alone in Yankee territory—affects your performance. If you perceive yourself as part of the in-group, and if that in-group is high performing, then you are more likely to perform well, just because you are part of that group. Alternatively, if you perceive yourself as outside the in-group that performs well, or if you perceive yourself as part of a group that is low performing, you are more likely to do poorly.
Marilynn Brewer, a social psychologist at Ohio State University, calls this idea “optimal distinctiveness.” Optimal distinctiveness means that not only do individuals want to belong to a group, but they also care about how unique they are simply because they are a member of this group. People seek group membership, and also group differentiation. That’s why Red Sox fans care not only about the Red Sox and being part of their fan base but also about how they’re different from the Yankees and their fan base.
Research shows that belonging in groups, as well as being differentiated by groups, influences the way people think about themselves. When people are members of groups, they start to define themselves in the terms of the group. They also start to shift their concerns and motivations from themselves as individuals to the group as a whole. The relevance of all of this for education is that students’ self-identification as members of high-performing or low-performing groups actually affects their own, individual performance. When students see themselves as part of a high-performing group—for example, a particular ethnic group, some defined “talented” group within the school, or the like—they are more likely to perceive themselves as high-performing and then to actually do well. Conversely, if the student perceives herself as a member of a “low-performing” group, she is more likely to perceive herself as a low-performing student, and then actually perform worse. This middle step—where individuals see themselves as having the characteristics of the group of which they are a part—is essentially self-stereotyping. And what the research shows is that it actually leads to behavior that confirms the stereotype.
Furthermore, research that Brewer and others have done shows that when conceptions of self change—when people start to see themselves as members of groups and evaluate their success on this ground—their ideas about what might be their duty to others shifts. This effect is particularly pronounced in high-performing groups. When individuals are in the high-performing in-group category, they not only perform better, but they also “develop a cooperative orientation toward shared problems.”27 In a classroom setting, this means that when students see themselves as part of the in-group, and that group is high performing, they tend to care more about problems that are shared by the whole group.
Thinking back to Mazur’s and Treisman’s classrooms, Brewer’s theory has a lot of relevance. Essentially, most of what Mazur and Treisman do is to create a classroom setting that treats all students as possessing “merit” and as being capable of succeeding. In short, they make everyone part of the in-group. All students perceive themselves as part of the group that is high performing and develop collaborative attitudes that foster even greater success than had they been working on their own. It’s not surprising, then, that not only do those individuals succeed, as the studies show, but they also develop mind-sets that facilitate cooperation and help each other develop shared learning goals—which raises the tide for all.
FOR THOSE OF YOU who may wonder if the work of Mazur and Treisman is applicable to K–12 teaching, Stanford University professor Jo Boaler’s research of the Railside School offers one such example.28 Railside is a public high school in the Bay Area of California, considered to be on “the wrong side of the railroad tracks.”29 Students come from homes with few financial resources, and many are English-language learners. Yet in Boaler’s five-year longitudinal research, students in Railside’s high school math program outperformed the nearby school’s wealthier students on standardized tests, even though Railside students entered as ninth graders performing far below the comparison students on those tests. Not only did Railside students achieve greater mathematical gains, but more of them enrolled in higher-level math courses and enjoyed math more. The achievement gap between students from the “better” side of the tracks was greatly narrowed. Students also gained what Boaler calls “relational equity,” which means they learned to collaborate by valuing the ethnic, gender, and social-class diversity of their peers. The success of Railside provides a perfect example of the concepts I’ve discussed in this chapter: the power of a growth mind-set, the dynamic impact of teamwork, and the value of drawing from diversity.
In her research, Boaler studied three California high schools: Railside, Greendale, and Hilltop. Railside’s student body was 40 percent Latino, 20 percent African American, 20 percent white, and 20 percent Asian Pacific Islanders; 30 percent of students qualified for free or reduced lunch. Greendale had a predominantly white student body (90 percent) and low poverty rates (only 10 percent qualified for free or reduced lunch). At Hilltop, a rural high school with a student body that was 60 percent white and 40 percent Latino, 20 percent of their students qualified for free or reduced lunch. Another useful statistic to note is that 30 percent of the students at Railside were English-language learners, whereas no students at Greendale and only 20 percent of Hilltop’s students were.30
Boaler studied approximately three hundred students in “traditional” math classes at Greendale and Hilltop and approximately three hundred students in “reform oriented”31 math classes at Railside in a four-year study, following their progress from ninth through twelfth grades. When the study began, ninth graders entering Railside tested well below the performance of Greendale and Hilltop students. At the end of the first year of the study Railside students had approached the test scores of Greendale and Hilltop students. By the end of the second year, Railside students had significantly outperformed Greendale and Hilltop students.32 At the end of the third year, Railside students continued to outperform Greendale and Hilltop students, although not to a statistically significant degree.33 At the end of the fourth year, achievement tests were not administered because a more selective group of students remained in math classes in all three schools.34 Nonetheless, in this fourth year of the study, the twelfth grade students at Railside were enrolled in advanced math classes at a much higher rate (41 percent of students) than were the students at the other two schools (27 percent).35 Railside students also enjoyed math more, with 71 percent responding positively on a second-year questionnaire, compared to 46 percent of Greendale and Hilltop students.36 What made this stunning achievement possible, especially given the ninth graders’ lower entering test scores and the untracked classrooms? Teachers at Railside engaged in the same practices as Mazur and Treisman yet for younger learners: developing a growth mind-set, working in collaborative groups, and using diversity as an asset for achievement.
To begin with, Railside did not engage in the tracking practices typical of most high schools, where students are slotted into remedial, regular, or honors-level math courses that predetermine and restrict their educational opportunities both during and after high school. At Railside, all entering ninth graders were assigned to algebra. Interestingly, the Railside students who benefited the most from this untracked heterogeneous approach were the students who tested highest when entering the ninth grade—which should alleviate concerns that these students would be “held back” by their peers who didn’t test as well as they did on the math assessments.37
Next, teachers at Railside maintained high expectations of their students and offered them challenging tasks—all the while emphasizing the importance of effort over ability and continually reminding students that they could accomplish anything if they put in the effort.38 This message was internalized by students at Railside. In the words of Sara, a ninth grader:
To be successful in math you really have to just, like, put your mind to it and keep on trying—because math is all about trying. It’s kind of a hard subject because it involves many things. [. . .] But as long as you keep on trying and don’t give up, then you know that you can do it.39
Railside teachers taught their students not only that they must persist in the face of struggle but also how—by using “multiple-ability treatments.”40 Multiple-ability treatments emphasize the importance of multiple abilities required for a task. When beginning an activity, Railside teachers would discuss with students the multiple abilities necessary to complete the group task (for example, reasoning, creativity, spatial-visual reasoning) and told students, “None of us has all these abilities; each one of us has some of these abilities.”41 Instead, students were told they needed to work together to accomplish their math tasks. Throughout the students’ four years at high school, teachers continued to remind them of the importance of effort and persistence for success in math class. This was mainly done through teachers’ emphasis on collaborative group work.
How does this group work operate? Analyzing six hundred hours of classroom-observation video, collected throughout the study, Boaler found that Railside students worked in small groups 72 percent of the time and presented to the class 9 percent of the time. Only 4 percent of class time was spent with teachers lecturing to students, with 9 percent of class time spent questioning students as a whole class. In stark contrast, in the traditional classrooms at Greendale and Hilltop, 21 percent of class time was spent lecturing, 15 percent questioning students as a whole class, while 48 percent of the time students worked individually in their books, and only for 0.2 percent of the time did students present to the class.42 Writes Boaler:
Enhanced student participation had a purpose of course. Rather than the learning by rote which was occurring in the traditional classrooms of Greendale and Hilltop, students at Railside were expected to justify their reasoning, which not only improved their mathematics understanding but also created respect for each other. A 9th grader, Jasmine, explained the multiple approaches of her math class and the expectation of justification by saying, “It’s not just one way to do it [. . .] It’s more interpretive. It’s not just one answer. There’s more than one way to get it. And then it’s like: “Why does it work?”43
At Railside, students were expected to justify their reasoning for their answers and to question each other. Teachers emphasized their expectation that students take collective responsibility for their group. To do so, students were taught about positive teamwork through “complex instruction,” a group-work strategy that incorporates roles (for example, team captain, facilitator, recorder reporter, resource manager).44 Teaching students how to collaborate in groups not only benefited student learning, but it also facilitated respect for the diversity of the range of strengths that students brought to the group. Interviews with Railside students highlight the respect that students had for each other and the responsibility they took for both their own and their group’s learning.
INTERVIEWER: Do you prefer to work alone or in groups?
AMADO (Year 1): I think it’d be in groups, ’cause I want, like, people that doesn’t know how to understand it, I want to help them. And I want to—I want them to be good at it. And I want them to understand how to do the math that we do.
LATISHA (Year 3): It’s good working in groups because everybody else in the group can learn with you, so if someone doesn’t understand—like, if I don’t understand but the other person does understand, they can explain it to me, or vice versa, and I think it’s cool.
ZANE (Year 2): Everybody in there is at a different level. But what makes the class good is that everybody’s at different levels, so everybody’s constantly teaching each other and helping each other out.45
Railside students did not refer to classmates as “slow” or “dumb,” as they did at Greendale and Hilltop. Rather they referred to some students as those who “don’t do their work.”46 This highlights students’ internalization of a growth mind-set, where effort rather than ability determines success in mathematics. What is perhaps most interesting about this heterogeneous and group approach is that not only did students achieve greater mathematical gains, but the achievement gap was narrowed, and students learned to value each other’s ethnic and cultural backgrounds, the “relational equity” I mentioned.
Just as Scott Page showed that a diversity of students produced the best solutions, the students at Railside benefited from the diversity of their classrooms. Because students were not tracked into ability groups, math classes were diverse both in the different academic strengths that students brought to the classroom and also in the ethnic diversity of the student body—40 percent Latino, 20 percent African American, 20 percent white, and 20 percent Asian–Pacific Islanders. When teachers emphasized the importance of growth mind-sets, collaborative group work, and multiple perspectives, students learned that success in math comes from applying these principles. This is illustrated in this interview with Ayanna and Estelle, Railside seniors in the fourth year of the study.
INTERVIEWER: What do you guys think it takes to be successful in math?
AYANNA: Being able to work with other people.
ESTELLE: Be open-minded, listen to everybody’s ideas.
AYANNA: You have to hear other people’s opinions, ’cause you might be wrong.
ESTELLE: You might be wrong ’cause there’s lots of different ways to work everything out.
AYANNA: ’Cause everyone has a different way of doing things, you can always find different ways to work something out, to find something out.
ESTELLE: Someone always comes up with a way to do it. We’re always like, “Oh, my gosh, I can’t believe you would think of something like that.”47
Boaler proposes that this “relational equity,” where students treat each other respectfully and fairly consider each other’s diverse perspectives, should be one of the goals of education. She argues that schools should “produce citizens who treat each other with respect, who value the contributions of others with whom they interact, irrespective of their race, class or gender, and who act with a sense of justice in considering the needs of others in society.”48 At Railside, as students learned to value multiple and diverse perspectives in the classroom, they also came to value the ethnic and cultural diversity of their classmates. Students’ work in collaborative groups at Railside influenced their social relationships, as illustrated by this interview with Robert and Jon, also seniors in the fourth year of the study.
ROBERT: I love this school, you know? There are schools that are within a mile of us that are completely different—they’re broken up into their race cliques and things like that. And at this school, everyone’s accepted as a person, and they’re not looked at by the color of their skin.
INTERVIEWER: Does the math approach help that, or is it a whole school influence?
JON: The groups in math help to bring kids together.
ROBERT: Yeah. When you switch groups, that helps you to mingle with more people than if you’re just sitting in a set seating chart where you’re only exposed to the people that are sitting around you, and you don’t know the people on the other side of the room. In math you have to talk; you have to voice if you don’t know or voice what you’re learning.49
Students in the diverse classrooms of Railside had higher mathematics gains, outperformed the students of Greendale and Hilltop, enjoyed mathematics, and enrolled in advanced math courses. The achievement gap among ethnic groups either narrowed or disappeared.50 By contrast, achievement disparities by ethnic group at the traditional Greendale and Hilltop schools remained throughout the five years of the study.
Railside is not the only school engaging in this approach. The San Francisco Unified School District, for example, is currently in its fifth year of a professional development initiative in which middle and high school math teachers throughout the district participate in a thirty-hour summer course, meet throughout the year, review student work, and watch videos of each other’s instruction. Just as teamwork is expected of students, math teachers in these cities also work in teams. In Seattle, former Railside teacher and current University of Washington research associate Lisa Jilk coaches math teachers’ implementation of the “complex instruction” approach in Seattle public schools. At Vanguard High School, a public school in New York City in which most students qualify for free or reduced-priced lunches, math teachers meet weekly to support each other’s use of group work with their diverse students in the school’s untracked, heterogeneous math classes.
RAILSIDE AND OTHER urban public schools in San Francisco, Seattle, and New York City serve as enlightening and hopeful K–12 examples of the same culture shift advocated at the university level by Mazur and Treisman. Through collaboration and innovative approaches that allow students to value effort over ability, each of these examples follows the research of Dweck and Tough in promoting a growth mind-set rather than a fixed mind-set. By learning how to work productively in diverse groups, the shift in student perceptions on subjects such as intelligence results in greater learning gains, positive attitudes, and students’ appreciation of the diversity of their peers. This echoes both Woolley’s research on turn taking, social sensitivity, and the role of women, and Page’s research on the value of diversity in groups and multiple perspectives. This is democratic merit in action—and what could be the beginning of a widespread culture shift that will ultimately benefit both individuals and society as a whole.