Students’ sense of belonging, and what we can do about it

Last week, Sarah Hammarlund (of “Context Matters: How an Ecological-Belonging Intervention Can Reduce Inequities in STEM” by Hammarlund et al., 2022) gave a presentation here at LTH as part of a visit funded by iEarth* that led to a lot of good discussions amongst our colleagues about what we can do to increase students’ sense of belonging, and to the question “what can we, as teachers, do, to help students feel that they belong?”.

Below, I’m throwing together some ideas on the matter, from all kinds of different sources.

But first, why does it matter whether students feel like they belong? In many studies, student sense of belonging (being defined in many different ways) has been linked to successful transition into university, persistence when faced with adversity, graduation rates at university, academic success in general, mental health, and many other desirable outcomes. So it is not just about whether students “feel good”, it is about giving them the best possible opportunities to succeed. And student sense of belonging is not distributed equally or randomly — low sense of belonging is often found in historically marginalized groups of people: women, people of color, first generation students. So by focussing on how we can help students feel that they belong, we can — to some extent — help level the playingfield for students that probably have already experienced more systematic challenges than others, and in general create a more welcoming and inclusive atmosphere.

How to support different components of students’ sense of belonging

I really like the model by Green et al. (2016) that divides “sense of belonging” into two components: Social belonging (with peers and teachers) and academic belonging (with the subject and school). Green et al. (2016) find that while there are students that report a high sense of belonging with both their peers and teachers as well as the school, there are also those that just feel like they belong based on their friend group, and others that like the education they are getting but are not connecting with peers or teachers. So to me it makes a lot of sense to look at peers, teachers, and academic belonging separately to see how each of these can be supported independent of the others (even though, in many cases, supporting one aspect is also supporting other aspects at the same time).

Sense of belonging with peers

Whether or not students feel that they are part of a community of peers might ultimately depend more on what happens outside of class (and IF things happen outside of class). But there are some things we can do to create community within our classes. For example, using group methods where students get the chance to connect with each other while working together (for example on creating shared artefacts). Also help students carry those connections into the world beyond school by, for example, structuring the tasks in a way that students have to interact outside of class, and by making sure that there are opportunities for students to talk about other things than “just” work, so they can also connect on a social level. There are tons of methods for this, for example letting students arrange themselves in a space according to different questions, or letting students introduce themselves with their “nerd topic”.

When creating those groups, it is important that the same people come together regularly enough so they can actually connect (so in a class with hundreds of students, don’t mix them into new groups every week, or if you do so have them in addition have one group that they meet with consistently). When assigning groups, also try to cluster minorities rather than stretching them across as many groups as possible (Stoddard et al., 2020; Nice podcast with the author of the study: When there is only one “token woman” in a group, for example, she is perceived as less important by her male peers. She will then participate less in the group, and receive less credit when she does. In groups where there is a majority of women, the behavior of the male participants towards the women changes, and that (not homophily or self-assessment) is what changes how much influence the women get, and how well they perform.

Getting groups started, it is helpful to support the establishing of rules. And when the groups are working, it’s always good to keep a close eye on what’s going on, and being aware of how group processes typically work and, if necessary, intervene.

Sense of belonging with teachers

Cooper et al. (2017; my summary here) show that it matters to students whether teachers make an effort to know students’ names (but not whether the teacher actually does know their name!). When students think the instructor knows their names, it affects their attitude towards the class since they feel more valued and also more invested. Students then also behave differently, because they feel more comfortable asking for help and talking to the instructor in general. They also feel like they are doing better in the class and are more confident about succeeding in class. It also changes how they perceive the course and the instructor: In the course, it helps them build a community with their peers. They also feel that it helps create relationships between them and the instructor, and that the instructor cares about them, and that the chance of getting mentoring or letters of recommendation from the instructor is increased.

Here and in other studies is about whether students feel that the teacher cares about them as a person. So make sure you make it explicit that you care about students!

I also think that it helps if the teacher is giving students an authentic idea of who they are by sharing a bit about themselves. That can be edited, of course, but I am a big fan of, for example, letting students see the physical space from which I am talking with them on zoom (and that can be a curated corner of a bigger room that they are not going to see), rather than a virtual background (even though that can also be a good option if there is some information there that I want to present or discuss). As a teacher, especially with younger students, we easily become “the face of the subject”. If we are human, they can decide whether they like us and whether we are potentially a role model. If we don’t show our humanity, they don’t even have the chance to decide that they don’t like us and don’t want to ever become like us. Sharing is caring? [Edit 29.8.2022: But sometimes we might share things that we think make us relatable, but that is not the message that we are necessarily conveying, or only if we are providing enough context. See this post]

And in a nutshell, I think it is always good advice to communicate high expectations, that we trust students will be able to meet them, and they have our support working towards it.

Academic sense of belonging

Academic belonging is about feeling that this school, this subject are right for the student. While some of that might really be about the fit between the subject and student interests, that is far from the whole story. Learning outcomes and examples do not appeal to everyone in the same way (Stadler et al., 2000), but can be tailored to be appealing to a wide range of student interests.

When students are in a setting in, or doing an activity for which a negative stereotype for their group (or groups, see intersectionality!) applies, they will need energy to deal with knowing that others might react to or perceive them based on that stereotype, and to counteract reinforcing that stereotype. This energy is not available for performance, which leads to systematic underperformance of those students. This is called the “stereotype threat” (Steele, 2011), and it is important that we make sure to not accidentally (or otherwise) activate stereotypes:

  • Textbooks and other materials can perpetuate (and activate) stereotypes; even “objective” subjects – like physics – are often not represented in a gender-neutral way (for example using “he” instead of she/they; disproportionate numbers of diagrams, illustrations, examples using men or examples typically associated with men; gender role stereotyping (“men as welders, women as nurses”)) (Taylor, 1979) — this is something we can, and should, pay attention to.
  • The physical environment influences who feels welcome and participates: Physical environments that do not activate stereotypes (for example in computer sciences having nature posters on the wall instead of featuring masculine stereotype Star Trek poster and video games) boosts female sense of belonging and interest in the subject to the level of their male peers (Cheryan et al., 2009)
  • It matters who gives an exam because it can activate stereotype threat (Marx & Roman, 2002; Marx & Goff, 2005)
  • Competent female experimenters (i.e. role models) protect women’s math test performance by influencing women’s self-appraised math ability, which in turn leads to successful performance on a challenging math test (Marx & Roman, 2002)
  • If you have to ask about demographics, do it after the test so you don’t activate stereotype threat (Danaher & Crandall, 2008)
  • Telling women about other women’s achievements or having them read about individual success stories helps them perform better and not be hindered by stereotype threat (McIntyre et al., 2003). This probably also holds for other minorities.
  • Make sure to include all relevant voices (“Decolonizing the curriculum”; Dessent et al., 2022; Some of my thoughts here:

In general, students achieve more when they believe that they can develop their abilities and overcome challenges through good strategies, hard work, learning new skills, help from others, and that it takes patience. Such a “growth mindset” helps them persist and show higher achievement in challenging situations. A growth mindset can be supported for example by modelling overcoming of challenges, and by pointing out that success is not due to traits, but rather due to effort (Yeager & Dweck, 2012). Even non-feedback related teacher comments, for example about a subject or a course in general (in the article, they give the examples of “not everyone is good at statistics” vs “everyone can learn statistics if they try”; and “sorry no time for questions” vs “please ask questions”), can influence students’ mindset towards a growth mindset, and thus influence their performance (Smith et al., 2018).

And there are many interventions, like the one in Hammarlund et al. (2022) that we are working with right now, that help shape classroom norms and instill the understanding that challenges are normal, temporal and surmountable. I would argue that this intervention, in which students first write about anticipated challenges and later discuss challenges and solutions, would work on all three components of belonging: Not just the obvious academic, but also both the belonging with peers through creating a community in which conversations about challenges and solutions are possible and normal, and belonging with the teacher, because they care enough to spend classtime on facilitating these kinds of discussions.

Is belonging really what we should strive for?

I recently listened to the Teaching in Higher Ed podcast “How Mattering Matters, with Heidi Weston and Peter Felten” and they point out the very important distinction between “belonging” and “mattering”. “Belonging” always includes an element of “fitting in” into a specific group of people or an institution. But there might be students who don’t want to “fit in”, for example because of the history of a specific institution or their personal history, yet they still want to learn from that very institution. I, for example, did not participate in any of the student initiation rituals when I started university, because I did not want to belong to a group based on activities that center around alcohol and dares (and I still find it super off-putting, when I see student hordes engaging in those initiation rituals). And I have definitely met my fair share of anti-role models. But I still wanted to be an oceanographer, and learn in the context of that university, and I still want to work at a university now. So a high degree of academic belonging for me, but a highly curated social belonging.

Another aspect to consider about belonging: when “belonging” is about “fitting in”, it is very easy to put responsibillity of “belonging” on the student and forget about the teachers’ and institution’s role, or the role that I think they should be playing. With “mattering”, it is more obvious that there needs to be some activity on the teachers’ & institutions’ part.

For students that, for whatever reason, don’t want to “belong” in the sense of “fitting in”, all the other aspects of belonging are still relevant, though, and they still need to feel that they matter. So maybe “mattering”, showing students that they matter, is really what we should focus on going forward?

*Sarah’s visit was funded by a “seed funding” grant from iEarth. For this project, Sarah gave several presentations at UiB and UiO and discussed with colleagues there. In the picture at the very top of this blog post, you see me taking the selfie, and from left to right, Emily Christiansen, Sehoya Cotner, Sarah Hammarlund, and Robin Costello working on coding responses to a survey we ran in undergraduate STEM courses at UiB and UiO to start working on topics related to sense-of-belonging and test-anxiety there.


Cheryan, S, Plaut, VC, Davies, PG, and Steele, CM (2009), Ambient belonging: how stereotypical cues impact gender participation in computer science, Journal of Personality and Social Psychology, Vol. 97, No. 6, pp. 1045.

Cooper, K. M., Haney, B., Krieg, A., & Brownell, S. E. (2017). What’s in a name? The importance of students perceiving that an instructor knows their names in a high-enrollment biology classroom. CBE—Life Sciences Education, 16(1), ar8.

Danaher, K, and Crandall, CS (2008), Stereotype threat in applied settings re- examined, Journal of Applied Social Psychology, Vol. 38, No. 6, pp. 1639-1655.

Dessent, C. E., Dawood, R. A., Jones, L. C., Matharu, A. S., Smith, D. K., & Uleanya, K. O. (2021). Decolonizing the Undergraduate Chemistry Curriculum: An Account of How to Start. Journal of Chemical Education.

Green, M., Emery, A., Sanders, M., & Anderman, L. H. (2016). Another path to belonging: A case study of middle school students’ perspectives. The Educational and Developmental Psychologist, 33(1), 85-96.

Hammarlund, S. P., Scott, C., Binning, K. R., & Cotner, S. (2022). Context Matters: How an Ecological-Belonging Intervention Can Reduce Inequities in STEM. BioScience, 72(4), 387-396.

Marx, DM, and Goff, PA (2005), Clearing the air: The effect of experimenter race on target’s test performance and subjective experience, British Journal of Social Psychology, Vol. 44, No. 4, pp. 645-657.

Marx, DM, and Roman, JS (2002), Female role models: Protecting women’s math test performance, Personality and Social Psychology Bulletin, Vol. 28, No. 9, pp. 1183-1193.

McIntyre, RB, Paulson, RM, and Lord, CG (2003), Alleviating women’s mathematics stereotype threat through salience of group achievements, Journal of Experimental Social Psychology, Vol. 39, No. 1, pp. 83-90.

Smith, T., Brumskill, R., Johnson, A., & Zimmer, T. (2018). The impact of teacher language on students’ mindsets and statistics performance. Social Psychology of Education, 21(4), 775-786.

Stadler, H., Duit, R., and Benke, G., “Do boys and girls understand physics differently?,” Phys. Educ.0031-9120 35(6), 417–422 (2000).

Steele, C. M. (2011). Whistling Vivaldi: How stereotypes affect us and what we can do. WW Norton & Company.

Stoddard, Olga B.; Karpowitz, Christopher F.; Preece, Jessica (2020) Strength in Numbers: Field Experiment in Gender, Influence, and Group DynamicsIZA Discussion Papers, No. 13741, Institute of Labor Economics (IZA), Bonn

Taylor, J. “Sexist bias in physics textbooks,” Phys. Educ.0031-912014(5), 277–280 (1979).

Yeager, DS, and Dweck, CS (2012), Mindsets that promote resilience: when students believe that personal characteristics can be developed, Educational Psychologist, Vol. 47, No. 4, pp. 302-314.

Leave a Reply