The WWC recently reviewed the strength of evidence for two types of interventions designed to help students succeed in college: one report focuses on growth mindset interventions and another on social belonging. The WWC found that (1) neither type of intervention had a discernible effect on full-time college enrollment, (2) social belonging interventions had mixed effects on progressing in college and academic achievement, and (3) growth mindset interventions had potentially positive effects on academic achievement. We asked Greg Walton, an Associate Professor at Stanford University, IES-funded researcher, and expert on these kinds of interventions, to discuss what college faculty, administrators, and students should make of these findings.
Can you walk through how growth mindset interventions and social belonging interventions with postsecondary students work? Were the interventions reviewed by the WWC typical interventions in this space?
Growth mindset interventions focus on the underlying “implicit” beliefs students have about the nature of intelligence: Is intelligence fixed or can it grow? These beliefs inform how students make sense of everyday academic challenges in school. If you think that intelligence is fixed, that you either have it or you don’t, then a setback like a poor grade can seem to be evidence that you don’t have what it takes. That can make students avoid academic challenges, withdraw, and ultimately learn and achieve less. Growth mindset interventions offer students the view that intelligence can grow with effort, hard work, good strategies, and help from others. The theory is that that mindset can help students see setbacks simply as evidence that they haven’t learned the material yet, or that their strategies haven’t been successful yet, and thus to sustain their efforts. These interventions typically start by sharing information from neuroscience about how the brain grows “like a muscle” during learning, especially when students work on challenging material. Then students might read stories from older students who used a growth mindset to persist through challenges. Finally, they may be asked to describe this idea to help younger students struggling in school, a technique termed “saying-is-believing.” That makes the experience active rather than passive and positions students as benefactors rather than beneficiaries, which would be potentially stigmatizing.
Social-belonging interventions target “belonging uncertainty,” a persistent doubt students can feel about whether “people like me” can belong in a school setting. This doubt arises most strongly for people who belong to groups that have historically faced exclusion in school settings, negative stereotypes that pose them as less able and less deserving of educational opportunities, or who are underrepresented in a school context. When students experience this doubt, everyday challenges such as feeling lonely, being excluded, or getting critical feedback can seem like evidence that they don’t belong in general. Social-belonging interventions share stories from older students who describe how they worried at first about whether they belonged in a new school and how these worries dissipated with time as they developed friendships and study partners, joined student groups, and formed mentor relationships. Belonging interventions offer students the view that it’s normal to worry about belonging at first in a new school but this gets better with time. Like growth mindset interventions, belonging interventions use written exercises to give students the opportunity to reflect on the intervention message and advocate for it to younger students. The theory is that this message can help students sustain a sense of belonging and stay engaged in school even when they face challenges, and that that helps students develop friendships and mentor relationships that support higher rates of achievement.
Social-belonging interventions were designed specifically to address circumstances in which people face underrepresentation or negative stereotypes in school. Even if all students have reasons to worry whether they belong in school, only some students have reason to question whether “people like me” belong. I am a White person whose parents both graduated from college. So, when I went to college, I felt homesick but I didn’t wonder whether “people like me” could belong.
That said, belonging concerns are felt by almost everyone, and in some cases belonging interventions have produced main effects (benefits for all students) rather than interactions predicated on group identity (e.g., Borman et al., 2019 for evidence from students in grade 6). However, most trials find greater benefits for students who face underrepresentation or negative stereotypes in specific settings. One study found that women in more gender-diverse engineering majors (averaging 33% women) showed no achievement gap with men in the first year and no benefit from a belonging intervention. But women in male-dominated majors (averaging 10% women) showed a large achievement gap in first year performance, but that gap was closed by the intervention (Walton et al., 2015; see also Binning et al., 2020) [Editor’s note: These two latter studies did not meet WWC standards for internal validity. Although this suggests caution in drawing conclusions from the studies, failing to meet WWC standards does not imply that an intervention is ineffective.]
Taken together, a fixed-mindset of intelligence and belonging uncertainty can be like a toxic tornado for students, swirling into each other and creating cascading self-doubt. I’m describing these interventions separately because they grew up independently in the literature, and the WWC’s two reports look at each separately. But for students, they are often experienced together.
It’s also important to state that, although the interventions reviewed by the WWC are typical of those conducted with postsecondary students, these are highly active areas with new trials reported regularly. Studies have explored new populations and college contexts (e.g., Murphy et al., 2020) and are increasingly focused on identifying boundary conditions that determine where we should and should not predict effects (see Bryan, Tipton, & Yeager, 2020). It is also noteworthy how few studies have examined the critical question of progress in college (3 in each report). We need much more research here, exploring effectiveness, implementation strategies, and boundary conditions. Further, research is increasingly complementing direct-to-student interventions by exploring how we can support practices in school that support growth mindset and belonging (Murphy et al., 2021). For example, recent research shows that highlighting pro-diversity peer norms—namely that most students endorse diversity—can facilitate more inclusive treatment among college students and, in turn, reduce achievement gaps between privileged and marginalized students (Murrar et al., 2020).
What are the key components that are needed for a social belonging or growth mindset intervention to have a good chance of working? What elements need to be in place to help students improve academically or to stay enrolled in college?
I would distinguish two layers of this question.
One layer is what does it take for a discrete exercise focused on belonging or growth-mindset—such as the focus of the trials reviewed by WWC—to help students. In general, we should consider what, how, when, and where.
What is it you want to offer students? It should give students an authentic and adaptive way to make sense of common challenges they face, a way of thinking they can use to achieve their goals in college. Simple exhortations such as, “I know you can do it” or “You belong!” do not effectively impart a growth mindset or a sense of belonging, as Carol Dweck and I have written. Instead, it is useful to use high-quality materials developed and validated in research. Examples of materials available online are here and here.
How will you convey this? The goal of these interventions is to address foundational beliefs students have about school, such as “Can I do it?” and “Can people like me belong here?” It’s not to do something else, like to build a skill. That means the experience need not take long—typically, interventions last 30-60 minutes—but it should be immersive and interactive. You want students to deeply reflect on the ideas you present and connect these ideas to their lived experience.
That said, the more you can implement approaches that are scalable within an institutional context the more students you can potentially help. That’s one reason recent trials that reach large samples have focused on online modules (e.g., LaCosse et al., 2020; Yeager, Walton, & Brady et al., 2016). Students can log-on individually and complete materials at near-zero marginal cost. However, these approaches also have challenges, as online modules may not be as engrossing as in-person experiences. As we have moved from delivering these interventions in one-on-one, in-person experiences to larger studies with materials delivered online, we have found that students spend less time on the same materials and write less in response to prompts. Another alternative is having students meet in-person in groups to participate in these interventions or discuss their content (see Binning et al., 2020; Murphy et al., 2020), but that may be more difficult to implement on a large scale. So, there can be trade-offs between reaching scale and creating deep and impactful experiences.
When should you do this? In general, it is valuable if an intervention happens earlier rather than later, so it can alter trajectories going forward. However, it may be optimal to deliver interventions soon after students have encountered some challenges, but before they have taken steps in response to those challenges that are hard to reverse (e.g., dropping out). In general, social-psychological interventions are more sensitive to timing than to dosage. Growth mindset and belonging interventions have been delivered from the summer before college (Yeager, Walton, Brady, et al., 2016), to the first academic term (Walton et al., 2015), to the second (Walton & Cohen, 2011).
Where should you deliver interventions? This brings us to the second layer. So far, I’ve addressed the first layer, where you are focused on a discrete experience or set of experiences. But the second layer is that, growth mindset and belonging interventions will be most effective in contexts in which (1) the message offered is legitimate and authentic (locally true) and (2) students have real opportunities to get academic support and to develop a sense of belonging. In the end, to produce the most robust change, we must create cultures in schools in which adaptive ideas about ability and belonging are normal and reinforced. There are many ways that institutions signal to students, even inadvertently, messages about the nature of intelligence and who belongs. In welcoming a new class to campus, do we extol the past achievements of a few, which may only heighten imposter syndrome among everyone else? Can we instead talk about what students can do in the future and who they can become? In welcoming students to class, do faculty communicate that they expect to weed out large numbers of students? Or do they design assignments and evaluations to support students’ learning and growth (Canning et al., 2019)? Another question involves how well colleges foster opportunities for students to develop in-group pride and identity. Tiffany Brannon at UCLA finds that African American students do better in college when they have more opportunities to participate in events that celebrate and explore Black culture (Brannon & Lin, 2021). Some resources to help researchers and practitioners create cultures of growth and belonging for all students are available at the Student Experience Project, co-led by the College Transition Collaborative (https://collegetransitioncollaborative.org/student-experience/).
Recently, you and your colleagues have distinguished between people with different characteristics - and environments with different characteristics. You’ve argued that researchers should be looking more closely at the contexts, or what you’ve called “psychological affordances” in which these interventions might have different effects. Why is this work important? Why should educators be paying attention?
Social-psychological interventions operate within complex systems. Those systems invariably determine the specific effect any intervention has. To understand this, my colleagues and I have found it useful to consider the affordances of a school context: What does a context make possible (Walton & Yeager, 2020)? For instance, no psychological intervention will help English-language speakers learn Chinese if they aren’t receiving instruction in Chinese.
We distinguish two kinds of affordances. One is structural: What is it that different institutions make possible for students to do? As an example, in a forthcoming study, Shannon Brady, Parker Goyer, David Yeager, and I tracked college outcomes of students randomly assigned to a social belonging intervention or a control condition at the end of high school. The intervention raised the rate of bachelor’s degree completion for students who first enrolled in more selective 4-year institutions from 26% to 43%. These are institutions that tend to have higher retention and graduation rates and tend to spend more per student on instruction and student services than less selective 4-year institutions. They thus afford higher 4-year completion rates. At the same time, the same belonging intervention had no effect on bachelor’s degree completion rates for students who first enrolled in less selective 4-year institutions.
The second kind of affordance is psychological: What is it that students can believe in a school context? Does the cultural context in which an intervention is delivered one in which the way of thinking offered by the intervention can take hold and thrive? Or is it one that makes that way of thinking illegitimate, inauthentic, or not useful? A large-scale social-belonging intervention delivered online to students in 21 diverse colleges and universities increased first-year full-time completion rates for students from historically underperforming groups, but only in colleges that afforded, or fostered, a sense of belonging to members of those groups. Let’s break this down: In some college contexts, students from historically underperforming groups (who were not exposed to the intervention) realized a high sense of belonging by the end of the first year. Here the belonging message was “locally true” (true here, for people like me). Although we don’t know exactly why this was the case, presumably in these schools students from the given group had more opportunities to develop friendships, to join student groups, and to form meaningful relationships with instructors. In other colleges, students did not attain this high sense of belonging by the end of the first year. Only in the first case did the belonging intervention raise first-year completion rates (Walton, Murphy et al., in prep; described in Walton & Yeager, 2020).
In both cases, the belonging intervention helped students take advantage of opportunities available to them, whether to graduate or to belong. An important implication is that it may be necessary to address both students’ beliefs and whether contexts support more positive beliefs. That’s helpful, because it gives us a precise way to think about how to make contexts more supportive: To what extent do they make adaptive beliefs about intelligence and belonging legitimate and authentic and, if they do not, what can we do about this?
It sounds like you’re saying postsecondary leaders who want to foster greater student success and reduce gaps in retention and academic performance may want to consider these kinds of interventions, in part because they are relatively inexpensive to deliver to large numbers of students. But they should also consider how hospitable their campus is to students who might initially struggle in college.
For example, to reinforce a growth mindset, universities need to make academic support resources truly accessible; to reinforce a sense of belonging, universities might look for multiple ways to communicate that successful students of all kinds of backgrounds have initially experienced self-doubt, and that feeling like you don’t belong is a fairly normal and temporary part of adjusting to college.
That’s right. Growth mindset and belonging are about both student beliefs or ways of thinking and institutional practices—either alone may not be enough. So, to support a growth mindset, institutions should both (1) convey that all students can learn and grow with effort, good strategies, and support from others and (2) back that up by creating learning environments designed to support growth, including adequate academic supports, and classes that focus on fostering growth rather than identifying who is allegedly smart and who is not. To support belonging, institutions should (1) acknowledge that nearly all new college students worry at first about whether they belong, that this is normal and improves with time and (2) create classroom and out-of-classroom environments in which all of the diverse students we serve can develop strong friendships and mentoring relationships and find communities in which they belong.
Thanks very much, Greg.
Read the WWC’s summary of evidence for these interventions in the Growth Mindset Intervention Report and the Social Belonging Intervention Report. Find related resources at the The College Transition Collaborative (https://collegetransitioncollaborative.org/) or the Project for Education Research That Scales (https://www.perts.net/)
Carter Epstein, Senior Associate at Abt Associates, produced this blog with Greg Walton, Associate Professor of Psychology at Stanford University.
Note: The discussion above reflects the opinions of Greg Walton and does not necessarily reflect the opinions of the Institute of Education Sciences or the What Works Clearinghouse. Some of the studies cited above have not been reviewed by the What Works Clearinghouse.
Binning, K.R., Kaufmann, N., McGreevy, E.M., Fotuhi, O., Chen, S., Marshman, E., Kalender, Z.Y., Limeri, L., Betancur, L., & Singh, C. (2020). Changing social contexts to foster equity in college science courses: An ecological-belonging intervention. Psychological Science, 31,1059-1070. https://doi.org/10.1177/0956797620929984
Borman, G.D., Rozek, C.S., Pyne, J., & Hanselman, P. (2019). Reappraising academic and social adversity improves middle school students’ academic achievement, behavior, and well-being. Proceedings of the National Academy of Sciences 116 (33), 16286-16291. https://doi.org/10.1073/pnas.1820317116
Brady, S. T., Walton, G. M., Goyer, J. P., & Yeager, D. S. (in prep). [Where does a brief belonging intervention increase the attainment of a college degree? The role of institutional affordances.] Manuscript in preparation.
Bryan, C. J., Tipton, E., & Yeager, D. S. (2021). Behavioural science is unlikely to change the world without a heterogeneity revolution. Nature human behaviour, 5(8), 980–989. https://doi.org/10.1038/s41562-021-01143-3
Bryk, A. S., Grunow, A., Gomez, L. M., & LeMahieu, P. G. (2015). Learning to improve: How America’s schools can get better at getting better. Harvard Education Press.
Canning, E. A., Muenks, K. ,Green, D.J., & Murphy, M.C. (2019). STEM faculty who believe ability is fixed have larger racial achievement gaps and inspire less student motivation in their classes. Science Advances, 5(2). https://www.science.org/doi/10.1126/sciadv.aau4734
Dweck, C. (2016, January 11). Recognizing and overcoming false growth mindset. Edutopia. https://www.edutopia.org/blog/recognizing-overcoming-false-growth-mindset-carol-dweck
Murphy, M.C., Fryberg, S.A., Brady, L.M, Canning, E.A., & Hecht, C.A. ( 2021, August 25). Global Mindset Initiative Paper 1: Growth mindset cultures and teacher practices. https://ssrn.com/abstract=3911594
Murrar, S., Campbell, M.R. & Brauer, M. (2020). Exposure to peers’ pro-diversity attitudes increases inclusion and reduces the achievement gap. Nature Human Behavior 4, 889–897 . https://doi.org/10.1038/s41562-020-0899-5
Walton, G.M. (2021, November 9). Stop telling students, “You belong!” Three ways to make a sense of belonging real and valuable. Education Week. https://www.edweek.org/leadership/opinion-stop-telling-students-you-belong/2021/11
Walton, G. M., Logel, C., Peach, J. M., Spencer, S. J., & Zanna, M. P. (2015). Two brief interventions to mitigate a “chilly climate” transform women’s experience, relationships, and achievement in engineering. Journal of Educational Psychology, 107(2), 468–485. https://eric.ed.gov/?id=EJ1061905
Walton, G. M., Murphy, M. C., Logel, C., Yeager, D. S., Goyer, J. P., Brady, S. T., . . . Krol, N. (in preparation). Where and with whom does a brief social-belonging intervention raise college achievement? Manuscript in preparation.
Walton, G. M. & Yeager, D. S. (2020). Seed and soil: Psychological affordances in contexts help to explain where wise interventions succeed or fail. Current Directions in Psychological Science, 29, 219-226. http://gregorywalton-stanford.weebly.com/uploads/4/9/4/4/49448111/waltonyeager_2020.pdf
Yeager, D. S., Walton, G. M., Brady, S. T., Akcinar, E. N., Paunesku, D., Keane, L., Kamentz, D., Ritter, G., Duckworth, A. L., Urstein, R., Gomez, E. M., Markus, H. R., Cohen, G. L., & Dweck, C. S. (2016). Teaching a lay theory before college narrows achievement gaps at scale. Proceedings of the National Academy of Sciences of the United States of America, 113(24), E3341-E3348. https://doi.org/10.1073/pnas.1524360113