Association for Public Policy Analysis & Management

The economic returns to a postsecondary education continue to rise (Autor, 2014) and the lifetime wage premium is different across majors. For example, STEM graduates receive on average $1.5 million more than high school graduates over their lifetime (Webber, 2014; Webber 2016). For decades education interventions focused on helping students to apply to college and improving remediation placement and/or college-level course-taking for less-prepared and disadvantaged students. However, still nearly half of students who begin college do not complete (Bound, Lovenheim, & Turner, 2010) and low-income students and underrepresented minority students remain less likely than higher-income and white peers to complete the postsecondary degree (Bailey & Dynarski, 2011), which further contributes to disparate outcomes (Bjorklund-Young, 2016; Dynarski, 2015).

One promising type of intervention that has the potential to meet these needs involves “light-touch” growth mindset interventions where treatment conditions entail both relatively low respondent burden and costs to administer (Walton, 2014; Yeager & Dweck, 2012; Yeager & Walton, 2011). In randomized studies these interventions have already demonstrated statistically significant positive outcomes on GPA and persistence, particularly effective for students from historically underrepresented in higher education (Broda, et al., 2018; Paunesku, et al., 2015; Yeager, et al., 2016). In this paper, we measure the impact of a mindset intervention on the academic performance and persistence at Michigan State University (MSU) and extend previous literature by investigating the mechanism through which the intervention may impact the degree completion.

The Spartan Persistence Project is a multi-year longitudinal study that uses a randomized control trial design. Beginning in 2014/2015, each cohort of all prospective MSU students has received an email with a link to an online survey about growth mindset in the summer before their first year of college. The email also randomly assigns prospective students to one of three conditions: 1) growth mindset intervention; 2) belonging intervention; and 3) no intervention (control). The students assigned to the growth mindset intervention read an article and completed an exercise that emphasized the message that intelligence can be developed with efforts.

We examined the degree to which the intervention has effects on the academic achievement and persistence over time by tracking 2014/2015 cohort over their first eight semesters and on the graduation status. Results indicate that the treatment has no impact on the probability to graduate within four years, academic persistence, and GPA. The intervention, however, improved the predicted probability to graduate with selective majors, especially for Pell-eligible, female, and Latino students. We investigated three student behaviors that are consistent with growth mindset: taking challenging classes, choosing a challenging major, and re-taking courses. Results show that the intervention increased the likelihood to take challenging courses during their first year of school. Mindset treated students were also more likely to opt into a challenging major during their second year of school and to graduate with the challenging major. Both of these effects were most pronounced for students from low-income backgrounds and Hispanic students. In addition, only mindset-treated Pell-eligible students were more likely to re-take failed courses.