< BACK TO K-12 Education

A Q&A with Character Lab's Dr. Angela Duckworth

Taking a closer look at the notion of "grit" and the future of character research

Character strengths like curiosity, grit, and optimism are not “extras.” These skills help students perform — in school, college, and life. The Walton Family Foundation is supporting Dr. Angela Duckworth, who popularized the notion of “grit” as an essential component of student success, and the Character Lab, which she founded with Dave Levin of KIPP and Dominic Randolph of the Riverdale Country School, to study innovative ways of teaching and assessing character. 

What inspired you to research character? 

Angela: Before I went into psychology research, I was a classroom teacher. I taught math, primarily to high school students but also in middle school. It just became clear to me — as I think it becomes clear to so many educators and parents — that whether kids are thriving or not that has a lot to do with their character, broadly defined. I, as a math teacher, was especially captivated by my students’ ability to exercise self-control and grit. 

What do we know about the relationship between students' character and their academic performance? 

Angela: Grit and self-control are tremendously powerful predictors of how well students of all socioeconomic backgrounds will do on everything from their report card grades to how often they’re going to get to class on time to whether they are going to graduate from high school or college, or, in fact, what they’re going to go on to do in terms of their professional success beyond school. And, of course, that’s the whole point of school, to put kids on the right path for life. It’s not an either/or. I think it’s a both/and. I think schools can absolutely encourage their students to develop character at the same time as they’re encouraging students to succeed academically, and if you do one, it should really help the other. 

And what don't we know yet? What are the top three questions you are trying to answer?

Angela: One major question — and maybe the most urgent one — is how do you increase character? The conversation on character development goes all the way back to Aristotle, and likely before. So, it’s not a new question. What’s new is that we are trying to use the scientific method. In other words, we’re trying to ask and answer the question in a more systematic way. I think we’re really just at the beginning. We know very little right now about how to systematically cultivate things like gratitude and growth mindset, intellectual curiosity, grit, self-control, and so on. We’re learning, but these are the first steps of a long journey. 

The second thing is measurement. You really can’t study anything and you can’t intervene in a systematic way if you can’t measure whether your program worked or didn’t work. For parents and educators, we want to create specific measures that provide useful feedback to children. Here’s an analogy: If a swimmer is trying to get faster, it’s really helpful to be able to look up at the clock and say, “That lap was one second faster than the last time.” We don’t have anything analogous so far for character. I think that’s an urgent priority. 

The third question is: once we do figure out how to intervene in effective ways and how to measure character with precision and accuracy, how do we scale any of this? How would we get this to the thousands of teachers and their many students in a way that is cost effective and high fidelity? That comes next after you solve the first two, but it’s on my mind already.  

What are the character traits that matter most in predicting student success? 

Angela: One of the most important things about character is that it’s plural. There are many ways in which children demonstrate that they can think, feel, and act to their own benefit and the benefit of others. That Aristotelian definition of character is embraced by the Character Lab and is also my definition of character. 

Achievement is one thing that is beneficial to youth, so we should focus on character strengths that predict that. Chiefly, those are things like self-control and grit — things that have to do with students’ own goals and their ability to organize their time and their energy to accomplish those goals. 

But there are two other major dimensions of character that I think merit at least as much attention. 

The second one is interpersonal character or “social character.” These character strengths include gratitude, empathy, social intelligence, generosity, kindness — all things that have to do with getting along in positive, productive ways with other people. It is possible to be high on self-control and grit and not be so high on these interpersonal strengths. 

The third, again, relatively independent, dimension of character is intellectual character. These are things like curiosity, critical thinking, a genuine love of learning, imagination. 

One reason I really like talking about a tripartite — a three dimensional — space for character is that it’s complete. If you have a kid who is able to organize his energy around his own goals and be effective, and who knows how and why to get along with others, and has a really fertile life of the mind, you have a pretty complete recipe for a child that’s growing up well. It’s not like there is an infinite list of character strengths that educators have to worry about. It’s really these three broad dimensions. 

Are there any types of people who exemplify the three dimensions? 

Angela: There are canonical gritty people, and canonical empathic people, and curious people, and then, of course, there are people who seem to be paragons of all three at once. I’m working on a book on grit, and I’ve had the pleasure of interviewing people like Kat Cole, the president of Cinnabon. She was raised by a single mom in Tallahassee, Florida. She worked her way up from being a Hooters waitress to being the president of a billion dollar international food conglomerate. She did that not just because she’s a smart lady, which she is, but also because she’s a paragon of grit. I should also add that she’s really a paragon of generosity, too. She sees it as her mission in life to bring out the best in people. She would say that intellectual curiosity is also one of her greatest strengths. So, I’d say she is an exemplar of character in the fullest sense of the word.

How good are students at assessing their own character strengths and weaknesses?

Angela: I think that by middle school, students are old enough to have some insight into their own character strengths and weaknesses. And there are some aspects of character that are very hard to observe from an outsider’s perspective, in which case you might defer to a child’s own judgment. Curiosity, for example, kind of happens inside someone’s head. But often teachers and others can be more accurate in assessing the behavioral aspects of character. One reason why teachers are especially valuable here is that kids’ standards aren’t always the same. Sometimes the kids who are hardest on themselves would be — by other people’s accounts — the most able or skilled in terms of character in the classroom. Then there are other kids who just think they’re great in almost every possible way and, in fact, teachers don’t agree with them — at least not as energetically. Those kids who are really off in their estimation, if you follow them over time, they do very poorly on objective measures of things like grades and test scores. 

Any assessment — including a self-report — is going to have limitations. For this reason, we want to explore other ways of measuring character. I think the ultimate idea would be to triangulate so you have a collection of measures, so you are able to compensate for one measure’s limitations with another. 

What sorts of technologies might help to measure character? 

Angela: One of the measures we’ve been working on in our lab is inspired by the marshmallow test. In the marshmallow test, a preschooler is told: “You can have this smaller amount now, or, if you wait, you can have this larger amount.” Universally, kids say, “I’ll wait. I’ll wait. I’d rather have the two marshmallows.” But the question is: how long they’re able to wait? That’s a very accurate measure of self-control. It was developed by Walter Mischel in the 1960s. That’s very hard to do in person, one-by-one with kids. It’s also hard to do with kids who are older than four, because they’re not as tempted by marshmallows and they can wait a lot longer. So there’s an engineering problem. 

Capitalizing on modern technology, we created kind of an electronic version of the marshmallow test. We also updated it with things older students these days would find a dilemma — a choice dilemma. In our test, which we call the academic diligence test, on the left side of the screen, there are math problems. They’re fairly simple — single digit addition, subtraction, multiplication. They’re really things kids can do, and we tell them: “Practice these math skills. It’s good for you to get these things on automatic, to get faster, no matter what level math you are.” And then we say, “By the way, if you want to, when you want to, at any time, you can take a break on the right side of the screen.” On the right side of the screen, there’s an ever-changing display of videos like movie trailers, music videos, and games like Tetris. 

When kids have the opportunity to play this game for 10 or 20 minutes, we assess how much time do they allocate to the math, which requires self-control but is good for them in the long run, and how much do they allocate to these entertaining electronic diversions, which feel good in the short run but are not very beneficial in the long run. We have found that the number of seconds students voluntarily allocate to math rather than fun things predicts things like their grades and how far they’re going to make it through college and is correlated with teacher ratings of grit and self-control and also self-reported ratings of self-control and grit. 

Have you performed that test on grownups? 

Angela: We haven’t. We’d first have to convince grownups that doing simple math problems is good for them. But if we could do that, I think it would correlate with their self-control. 

How does character fit into the broader conversation about measurement and accountability? 

Angela: We’ve had standardized tests for years. Like them or hate them, they are useful, I think, to do the job they’re supposed to do, which is to measure certain aspects of what kids know and are able to do. 

When you say character matters too, there’s a natural inclination to start assessing character in the same way and to have analogous teacher accountability policies, and so forth. I think that would really be going beyond what the measures can do right now. The thing that you don’t have for are non-fakeable assessments of character that are really apples-to-apples across school communities. I wrote a paper called “Measurement Matters” with my colleague David Yeager. And we did so because the White House asked us to. The White House said, “Everyone’s talking about measurement and character. Maybe someone should tell educators what the state of the art is and what are the promises but also what are the perils.” When the White House asks you, you say, “Yes,” so we wrote that paper in response. I do think it would be premature to have requirements that people start measuring grit in the classroom and tie teacher performance to it. We’re really not there. 

What do most educators do today to teach character? 

Angela: Many, many teachers will talk about encouraging character in some way, shape, or form in the classroom. Most teachers are going on intuition. I’ve heard of teachers who — on their own — have come up with things that scientists have studied like “gratitude letters.” There’s research showing that when you write a letter to someone, thanking them sincerely for doing something for you, that has all kinds of benefits. Some teachers, not even knowing about that research, come up with similar things on their own. I’m sure there are a lot of good things going on that I don’t know about that aren’t widely known. One of the problems with education, however, is that there is innovation going on, but it tends to be hermetically sealed behind the classroom door. A second problem is that we rarely test these ideas using the scientific method, so it’s really hard to know which of these teacher-generated ideas are useful and which aren’t.

One thing we’re doing to learn more about what teachers are doing, what works, and how we can scale it, is we are giving grants to teachers. Teachers probably have better ideas than we do about children and how to help them. What they don’t always have is training in the scientific method, measurement, study design, and statistics. What we’re hoping to do is help those teachers test those ideas in ways that might be more systematic than they might be able to do on their own. 

If you had a crystal ball, what would you predict teaching character would look like 10 years from now? 

Angela: My optimistic crystal ball would see that there are developed options that have real scientific research behind them. We will say to teachers from elementary school, if not earlier, all the way through high school, if not later: if you’re helping kids learn self-control, try this, if you feel like gratitude is important for your school community, try this. In addition to thinking about standalone interventions, we need to try to figure out how to get teachers to infuse character into their everyday practice. In the most successful cultures — if you look at high-performing military culture or sports culture — things are integrated. 

Are there any tips you give parents about how they can teach these skills at home? 

Angela: I’m a scientist and a character educator, but I’m also a mom, and so I think about this question a lot. First and foremost, a parent is a role model. I can’t emphasize this enough. I will give one little finding from the research that illustrates this.

Decades ago, a researcher named Al Bandura at Stanford did what were called later the Bobo doll experiments, where he had children watch grownups play with some toys for about a minute. In some cases, children then saw the grownup get up and take this three-foot tall inflatable childlike looking doll and then violently aggress toward it — kick it around and hit it with a hammer and punch it in the face and say awful things to the doll. The children weren’t given any explanation. Later on, the children were brought into the same room with the same toys and they were able to do what ever they wanted. The kids who had just watched the adult play normally with toys, did just that. They played with the toys and the session ended, and there was really nothing to say. The kids who had watched the adult beat up this doll, with hair-raising accuracy, mimicked exactly what they had seen the grownup do before them. It was an important demonstration of what many of us know intuitively: children, whether they are our students or our own kids, are watching us and they are imitating us. The first and foremost thing adults can do to encourage character is to model character. When you say to kids, “Work hard” and then you are the teacher who didn’t finish grading their tests because you were tired, how much are they going to believe that? Or if you say to your kids be nice to other people, but you’re not nice, they’re not going to model that.

The second thing I would say is that there’s a very rich literature on parenting, which I think can be summarized as this: it used to be thought that you could be a supportive parent — super warm and respecting your child’s individuality — or you could be a demanding, strict, high standards parent. And you had to pick somewhere on the continuum. It actually turns out that there isn’t a tradeoff. You can be extremely supportive, warm, tell them that you love them unconditionally, and, at the same time, be a very firm, high standards disciplinarian. There is not one continuum; there are two. This combination of being very warm and very demanding is called authoritative parenting. We don’t yet have the research linking authoritative parenting to character itself, but from everything we do know, the parents who are not only modeling character for their kids, but bringing up their kids in a supportive, warm, consistently disciplined, high expectations environment, are helping their kids stand the best chance of developing character as they get older.