Network researchers are currently fielding the largest-ever randomized controlled trial of growth mindset interventions in the United States. This study was designed to understand which kinds of students, in which kinds of classrooms, and which kinds of schools are most likely to benefit from online exercises designed to foster learning mindsets.

Lead investigator:

Collaborators and advisers:

The effort is supported by:

Primary Research Questions

  • What types of academic inequalities are best addressed by these interventions?
  • What are the school resources and school climates that make these interventions more or less effective?
  • What specific classroom resources and climates are necessary for the success of these interventions?

Why It’s Important

The study is one of the only longitudinal  studies of adolescent behavior outcomes in the history of the social and behavioral sciences to use the gold standard for testing cause and effect (a randomized experiment) with the gold standard for making claims about a population of schools (a random sample). Therefore, the results of the study can be generalized to the entire population of high schools across the United States.

At the end of the study, the team will have developed and tested a learning mindset program that can be delivered in any regular U.S. public high school, and they will know which kinds of schools and which kinds of students will benefit the most. At this point, the program will be offered to schools at no cost through scaling partners.

The researchers have also taken steps to ensure that the full dataset from the study will be made freely available to any scientist who wishes to analyze it; given the study’s rare design and comprehensive collection of student-, classroom-, and school-level measures, this dataset will be an invaluable resource that can provide countless insights about learning mindsets and the learning environment for years to come.

Intervention Being Tested

Students receiving the intervention learn about how the brain grows new connections between neurons as a result of effort and improved problem solving strategies, and the implications for students’ potential to become more intelligent through study and practice. Students also learn about how other students want to do well in school so they can have a career they enjoy and make a positive contribution to the world. Students internalize these messages through a series of writing exercises, in which they summarize what they’ve learned and advise a discouraged peer or future students using this knowledge. Students receive the intervention in two separate online sessions, which take about an hour to complete altogether.

Study Design

Researchers will compare the academic outcomes of a group of 9th graders who receive the mindset program with those who received a “placebo” (in this case, an identical online program that does not teach students about mindsets). The study is a double blind randomized control trial—the “gold standard” in scientific research. Because they are randomly assigned, the two groups should be identical in all characteristics. This means that any difference observed in outcomes between the treatment and control groups can be attributed to the program itself. Analyses will compare course grades, pass rates, retention, absences, disciplinary incidents, school dropout, and 10th grade course selection.


The study features a nationally representative sample of 76 regular U.S. public high schools; all 9th grade students enrolled in these high schools during the 2015-16 school year participated in the study (roughly 20,000 students). This type of probability sample allows us to infer that the results observed among our sample should generalize to the population from which our sample was chosen.

What This Study Can Tell Us

The results will show an “average effect” of the mindset program on student outcomes that is generalizable to the entire population of 9th grade students enrolled in regular high schools in the US. The study will also reveal that individual and contextual factors that influence the outcomes. For example, we will be able to say that the intervention is twice as effective in certain types of schools or classrooms as compared to the average.