There is now ample evidence that socio-emotional skills are as important as basic cognition in driving individual success. Among these skills, grit clearly stands out. Grit is generally defined as perseverance in a productive task in the face of — often repeated and frustrating — setbacks. While it has been documented that grit predicts a large array of achievement outcomes, until recently, very little was known about how malleable it was in childhood and, whether it was possible to foster it in the classroom environment. Our new paper provides convincing evidence in this regard.

To answer the question of “malleability” in a causal manner, we conducted a randomized field experiment in Turkey with almost 3000 fourth-graders (8-10 year olds) in over 100 classrooms. We chose to run this experiment in socioeconomically disadvantaged neighbourhoods in Istanbul, with the hope of addressing the widening socioeconomic achievement gap in the world.

We offered elementary school teachers from randomly chosen schools a specific curriculum promoting a growth mindset and encouraging teachers to adopt classroom practices that are consistent with such mindset. Growth mindset, in a nutshell, is a belief that skills can be improved through sustained effort. People who maintain this mindset are expected to persevere more, exhibit higher resilience to setbacks, handle frustration caused by failures and eventually reach their target. This hypothesized productive cycle is what motivated the paper’s colourful title, Ever Failed, Try Again, Succeed Better.

Image: Poverty Action Lab

Students received a minimum of 10 sessions of activities suggested by the curriculum. The curricular material emphasized: i) the plasticity of the human brain rather than the notion that ability is fixed and innate; ii) the role of effort in enhancing skills and achieving goals; iii) the importance of interpreting failures in a constructive way, something to learn from; and iv) the importance of goal setting. Students were encouraged to set goals, exert effort to work towards these goals, and to avoid interpreting immediate failures as a lack of innate ability or intelligence.

The intervention, however, was not simply delivering this curriculum. Rather, it was a pedagogical intervention, which aimed to change the classroom practices of teachers by changing their mindset. In this sense, the intervention was much more intensive than other growth mindset interventions typically implemented by psychologists.

The study offered a methodological innovation. Rather than using standard survey measures, we developed and used an incentivised real effort task to measure grit in children. This is how we measured it:

The students were given a grid of numbers and asked to find pairs of numbers that summed to 100. They were given a choice between a difficult and an easy version of the grid. They were also given a performance target, that is, to be successful in their chosen version they had to find at least three pairs in 1.5 minutes. The task was designed to have 5 rounds with each round providing students a new grid. To measure perseverance, students were given direct feedback by the experimenters after each round. For this, after a round is completed and before the choice for the next round is made, the experimenters marked students’ sheets by circling either “succeeded” or “failed”. The easy task yielded 1 prize in the case of success, and 0 for failure. Almost everyone attempted to do the easy task succeeded. The hard task yielded 4 prizes in the case of success, and 0 for failure, with only about 20 per cent of students reaching the performance target. These prizes were small gifts such as colourful stationary items and toys that are valuable to children.

After the completion of five rounds, students were informed that they were given an opportunity to play this game (their chosen version) exactly one week later. The significance of this component of the measurement is that all students were offered exercise sheets, which contain a large number of grids. The purpose of this was to measure whether students would seize the opportunity to improve their ability in this task using these sheets, and perform better in the following week.

We evaluated the intervention with respect to this behavioural outcome as well as math and verbal test scores. We conducted these tests as well as the aforementioned behavioural task by visiting classrooms in the absence of teachers.

Results suggest that the students in treatment schools were more likely to opt for a difficult, high-reward task. They were also more likely to re-attempt the difficult task after receiving a negative performance feedback. Also, these students were more likely to succeed at the challenging task during the second visit, suggesting that they exerted more effort to develop the task-specific ability, likely by using the offered practice materials.

What is perhaps more important from the education policy perspective is that the intervention increased math and verbal test scores by about 0.30 and 0.13 standard deviations respectively in the short term. A follow up after 2.5 years revealed that treated students’ math performance remained about 0.20 standard deviation higher than control students. The effect on verbal test scores dissipated in the long-term.

Given all these positive results, we also posed the question of optimality. Can one have too much grit? Specifically, is it possible that the intervention somehow disadvantaged the treated students by pushing them to try the challenging task? We answered this question by exploiting the incentivised nature of their behavioural task. We found that treated students finished up with more prizes implying that enhancing their grit and encouraging them to try and re-try was the optimal action within the context of this experiment.