Stanford psychologist Carol Dweck, along with other education researchers interested in growth mindset, have done numerous studies showing that when students believe their intelligence can grow and change with effort, they perform better on academic tests. These findings have sparked interest and debate about how to encourage a growth mindset in students both at home and at school.
Now, a national study of tenth-graders in Chile found student mindsets are correlated to achievement on language and math tests. And students from low-income families were less likely to hold a growth mindset than their more affluent peers. However, if a low-income student did have a growth mindset, it worked as a buffer against the negative effects of poverty on achievement.
“This is not a sample; this is everyone in school,” said Susana Claro, a doctoral candidate at Stanford's Graduate School of Education and lead author of the article “Growth Mindset Tempers the Effects of Poverty on Academic Achievement,” published in the Proceedings of the National Academy of Sciences. Claro, along with Stanford scholars David Paunesku and Carol Dweck, wanted to know if at a very large-scale (168,000 students) growth mindset would correlate with academic performance. They found that it did at almost every school in Chile, a correlation stronger than they expected to find.
When students in Chile take national exams measuring language and math, they are also obligated to fill out a lengthy survey from the Ministry of Education on a range of subjects, from bullying to healthy eating, sports and how well they liked their teachers. The survey questions change every year, and in 2012 Claro convinced the ministry to include two questions on growth mindset. Teachers and parents are also surveyed, which is why Claro and her colleagues have such detailed income information for each student.
“This is the first time that we can see the landscape of growth mindset in a complete population,” Claro said. “We’ve always been using samples before.” She and her colleagues wanted to know if a study this large would reveal the same correlations seen in representative samples in the U.S. or whether the large sample size would be “too noisy.” To Claro’s surprise, the findings were very clear.