How Children Succeed

impossible to change. William James says our nervous systems are like a sheet of paper. You fold it over and over and over again, and pretty soon it has a crease. And I think that’s what you at KIPP are doing. When your students leave KIPP, you want to make sure they have the kind of creases that will lead them to success later on.”

According to Duckworth, conscientious people don’t go around consciously deciding to act virtuously all the time. They’ve just made it their default response to do the “good” thing, meaning the more socially acceptable or long-term-benefit-enhancing option. In any given situation, the most conscientious path is not always the smartest option. On Carmit Segal’s coding-speed test, for example, the students who scored highest worked really hard at a really boring task and got nothing in return. One word for that behavior is conscientious. Another is foolish. But in the long run, it serves most people well to have conscientiousness be their default option. Because when it does matter—when you have to study for a final exam or show up on time for a job interview or decide whether to yield to temptation and cheat on your wife—then you will probably make the right choice, and you won’t have to exert yourself and exhaust yourself in order to do so. Strategies like MCII, or the act of imagining a picture frame around a marshmallow—in the end, these are just tricks to make the virtuous path easier to follow.

15. Identity

When I visited KIPP Infinity in the winter of 2011, halfway through the inaugural year of the character report card, character language was everywhere. Kids wore sweatshirts with the slogan “Infinite Character” and all the character strengths listed on the back. One pro-self-control T-shirt even included a nod to Walter Mischel: “Don’t Eat the Marshmallow!” The walls were covered with signs that read got self-control? and i actively participate! (one of the indicators for zest). There was a bulletin board in the hallway topped with the words Character Counts; tacked on the board were Spotted! cards, notecards that students filled out whenever they noticed fellow students performing actions that demonstrated character. (Jasmine R. cited William N. for zest: William was in math class and he raised his hand for every problem.)

I asked David Levin about the message saturation. Didn’t he think it was a little much? Not at all, he replied. “In order to succeed,” he explained, “this 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.”

Wall-to-wall messaging is nothing new at KIPP, of course; right from the start, Levin and Feinberg used posters and slogans and signs and T-shirts to create a powerful school culture at KIPP, to instill in students a sense that they were different, and that they belonged. Duckworth told me she thinks that KIPP’s approach to group identity is a central part of what makes the schools effective. “What KIPP does is create a social role shift, so that a child will suddenly switch into a totally different mindset,” she said. “They play on the in-group/out-group thing: ‘We know what SLANTing is and you don’t know what SLANTing is, because you don’t go to KIPP.’”

Psychologists have demonstrated that group identity can have a powerful effect on achievement—both a positive and a negative one. In the early 1990s, Claude Steele, a psychologist who is now the dean of the school of education at Stanford University, identified a phenomenon that he called stereotype threat. If you give a person a subtle psychological cue having to do with his group identity before a test of intellectual or physical ability, Steele showed, you can have a major effect on how well he performs. Researchers have since demonstrated this effect in countless different settings. When white students at Princeton were told before trying a ten-hole mini golf course that it was a test of natural ability in sports (which they feared they didn’t possess), they scored four strokes worse than a similar group of white students who were told it was a test of their ability to think strategically (which they were confident they did possess). For black students, the effect was the opposite: when they were told the mini golf course was a test of their strategic intelligence, their scores were four strokes worse. Steele’s theory is that when you are worried about confirming a stereotype about your group—that white people aren’t athletic; that black people aren’t smart—you get anxious, and as a result, you do worse.

Other researchers have found stereotype threat in pursuits much more serious than miniature golf. When people in their sixties and seventies and eighties were instructed to read an article about how memory fades with age before they took a memory test, they remembered 44 percent of the words in the test; members of a similar group who weren’t told to read the article before the test remembered 58 percent of the words. Before a challenging math test, female college students need only be reminded that they are female for them to do worse on the test than female students who don’t receive that identity cue.

The good news about stereotype threat is that, just as it can be triggered by subtle cues, it can be defused by subtle interventions. One of the most effective techniques, which has now been tested in a variety of settings, is exposing students at risk of stereotype threat to a very specific message: that intelligence is malleable. If students internalize that idea, these studies show, they gain confidence, and their test scores and GPAs often rise too.

The most intriguing fact about these interventions is that the question of the malleability of intelligence is actually hotly debated by psychologists and neuroscientists. Although scores on achievement tests like the SAT can certainly be affected by training of different kinds, the purest kind of intelligence is not very malleable at all. But a psychologist at Stanford named Carol Dweck has discovered a remarkable thing: Regardless of the facts on the malleability of intelligence , students do much better academically if they believe intelligence is malleable. Dweck divides people into two types : those who have a fixed mindset, who believe that intelligence and other skills are essentially static and inborn, and those who have a growth mindset, who believe that intelligence can be improved. She has shown that students’ mindsets predict their academic trajectories : those who believe that people can improve their intelligence actually do improve their grades.

And whether or not intelligence is malleable, mindset certainly is. Dweck and others have shown that with the right kind of intervention, students can be switched from a fixed mindset to a growth mindset, and their academic results tend to rise as a result. Joshua Aronson, a frequent collaborator with Claude Steele, and two colleagues conducted a study that compared the effectiveness of a few different mindset-changing interventions on a group of mostly low-income seventh-grade students in Texas. Over the school year, each student in the study worked with a mentor, a college student who met with him or her twice for ninety minutes each time and then communicated with him or her regularly by e-mail. Some students were randomly assigned to hear from their mentors a growth-mindset message such as “Intelligence is not a finite endowment, but rather an expandable capacity that increases with mental work.” Students in a control group heard a more standard message about the way that drug use could interfere with academic achievement.

At the end of the year, Aronson and his colleagues compared the two groups’ scores on Texas’s standardized achievement test, the Texas Assessment of Academic Skills, and the students who had heard a growth-mindset message did significantly better than the students who had heard the anti-drug message. The most impressive effect was seen in the math scores of the female students. The effect of stereotype threat has been well demonstrated in the math scores of girls and women, who seem to be especially anxious in testing situations when they think they might confirm the stereotype that girls are bad in math. In the Texas experiment, girls who received the standard anti-drug message averaged 74 on the test, about eight points below the male students who had heard the same message. The girls who heard a growth-mindset message averaged about 84, closing the gap with the male students completely.

16. Report Cards

Dweck’s notion that students do better when they think they can improve their intelligence applies to character as well. At least, that is the idea behind the character report card—that presenting character to students not as a set of fixed traits but as a series of constantly developing attributes will inspire them to improve those traits. I talked about this idea one morning at KIPP Infinity with Mike Witter, a thirty-one-year-old eighth-grade English teacher who seemed hard-wired to believe in the growth mindset. “If you’re going to be a good teacher, you have to believe in malleable intelligence,” he told me. “And character is equally malleable.