"I am not just another number": Do Engineering Teaching Assistants Feel Like They Belong, and Are They Representative?

Research Paper Recommended citation: de Lima, J., & Isaac, S. (2025). “I am not just another number”: Do Engineering Teaching Assistants Feel Like They Belong, and Are They Representative?. In Kangaslampi, R., Langie, G., Järvinen, H.-M., & Nagy, B. (Eds.), SEFI 53rd Annual Conference. European Society for Engineering Education (SEFI), Tampere, Finland. DOI: 10.5281/zenodo.17631802. This Conference Paper is brought to you for open access by the 53rd Annual Conference of the European Society for Engineering Education (SEFI) at Tampere University in Tampere, Finland. This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License. “I am not just another number”: Do Engineering Teaching Assistants Feel Like They Belong, and Are They Representative? J. de Lima a,1, S. Isaac b EPFL, the Swiss Federal Institute of Technology in Lausanne, Switzerland a 0000-0001-9235-9704 b 0000-0002-1527-8510 Conference Key Areas: Diversity, equity and inclusion in our universities and in our teaching Keywords: Belonging, teaching assistants, representation, first-year engineering ABSTRACT Sense of belonging is the subjective feeling of being valued by the group, and has been linked to students’ academic achievement, especially in STEM, and also to students’ demographic characteristics. Demographically similar near-peers (e.g. teaching assistants, TAs) have an elevated influence on students' sense of belonging and therefore their academic success, which makes ensuring that TAs are representative of the students in high attrition programs an important point for equity. This study explores the demographic representation and sense of belonging in 244 TAs recruited to assist in high-enrolment first-year courses in physics, maths and computing at an engineering university. We found the TAs are demographically representative of the student body for many aspects (gender, sexual orientation, migration background) but not for first generation status and country of previous studies (migration background). TAs reported a high sense of belonging to the institution, their departments and their disciplines, mostly without correlation to aspects of their identity. Specific experiences that affected their sense of belonging featured themes related to social inclusion and academic success. These findings are relevant for improving the academic experience of all students, most particularly those facing additional barriers to success. 1 [email protected] 1 INTRODUCTION Sense of belonging is the subjective feeling of being valued and supported by a group, engendering feelings of inclusion and a sense of identity related to the group. These feelings are dynamic, subject to influence from a desire to belong and perceived acceptance from the group (Mahar et al., 2013). Students’ sense of belonging in higher education has been found to play an important role in academic success and persistence. Several of the aspects previously found to influence students’ belonging, including academic recognition and social integration, have the potential to come together for students who become teaching assistants. This study examines the sense of belonging in a group of students recruited into their first TA roles and the alignment of their demographic profiles with the students for whom they are near-peer role models. This exploration is important to assess how well TAs are able to support the academic experience of all students, most particularly those facing additional barriers to success. Historically, many of the studies that explored belonging focussed on student behaviours and attributes that contributed to their social and academic belonging (Tinto, 1975). However, recognising the limitations of this approach, especially for diverse populations including minoritized students, students with disabilities, firstgeneration students, and students with diverse gender identities, multiple other frameworks have been proposed (Vaccaro & Newman, 2022). Additionally, current researchers also explore the influence of intersectional identities as well as institutional, community, and larger social factors on students’ sense of belonging (Bentrim & Henning, 2023; Cohen & Viola, 2022; Kahu et al., 2020; Lee et al., 2024; Museus, 2014; Museus et al., 2017; Strayhorn, 2018). In higher education, an increased sense of belonging is related to better mental health and emotional engagement (Jensen & Cross, 2021; Wilson et al., 2015), student persistence (Museus et al., 2017), improved academic performance (Schinske et al., 2016), student persistence (Museus et al., 2017; Seymour & Hunter, 2019) and improved academic self-efficacy (Cooper et al., 2017; Freeman et al., 2007), especially for underrepresented students including first-generation students (Harackiewicz et al., 2014), and students with lower incomes (Hansen et al., 2024). Factors that adversely affect a sense of belonging include demographic factors such gender, being an immigrant, or having parents who do not hold university degrees. Additionally, being a target of discrimination and harassment can increase feelings of exclusion. This can result in student attrition from STEM or dropping out from university (Seymour & Hunter, 2019). Belonging is not an immutable characteristic of the student, it arises from their lived experiences and is therefore malleable (Cohen & Viola, 2022; Kahu et al., 2020; Seymour & Hunter, 2019; Strayhorn, 2018). Belonging has been shown to be positively influenced by instructor behaviours (Cooper et al., 2017; Freeman et al., 2007), general pedagogical interventions (Cohen & Viola, 2022; Harackiewicz et al., 2014), as well as interventions specifically targeting students’ sense of belonging (Good et al., 2012; Schinske et al., 2016). Among the strategies that have been shown to have a positive impact on student sense of belonging and therefore retention, are providing opportunities for studentemployment including as teaching assistants (TAs) (Giacalone & Perrelli, 2022; McClellan et al., 2023). In STEM, such programs are also beneficial to the mentees/tutorees (especially minoritised students), and lead to increased academic performance, facilitate integration into campus community, persistence and retention (Akinla et al., 2018; Anderson et al., 2019; Knepper, 2019; Zaniewski & Reinholz, 2016). While mentors, such as TAs, prioritise their students’ personal wellbeing (de Lima et al., 2023, 2024), their influence on students’ academic achievement and sense of belonging, and their ability to be effective role models, is enhanced when they are demographically similar to their students (Atkins et al., 2020; Dennehy & Dasgupta, 2017; Estrada et al., 2022; Muir, 2018). As diverse populations are influenced by differing factors, it is important to implement multiple strategies that promote student success and to mitigate potential barriers. To be efficient in the choice and implementation of these strategies, institutions need to understand the factors that affect students’ sense of belonging on their campuses. In the past few years, the Swiss Federal Institute of Technology (EPFL), a large engineering school located in Switzerland, has worked to understand these barriers. A recent campus survey revealed that 40% of the student respondents did not feel like they were a part of the EPFL community (Le Duc, 2022; Tormey, 2021). Additionally, a recent report showed that the percentage of first-generation students at EPFL has dropped over the past 15 years, and that female first-generation students are less likely to choose EPFL than peers whose parents hold university degrees (Gerster et al., 2023, 2024). At a national level, a recent OECD report showed that there was a 25-percentage point difference in sense of belonging between native Swiss students and immigrant students (OECD, 2018). It is therefore essential to examine the demographic profiles of respondents in studies of belonging in higher education (Dias-Broens et al., 2024). EPFL has implemented several initiatives to address these barriers by helping firstyear students with their academic transition, especially for minoritised students, including first-generation and women students. One long-standing programme is a near-peer teaching assistantship programme where students in senior bachelor and Master levels are trained and employed to assist in the large enrolment first-year courses. As near-peer mentors’ capacity to support academic achievement and sense of belonging is stronger when they are demographically similar to their students (Atkins et al., 2020; Dennehy & Dasgupta, 2017; Estrada et al., 2022; Muir, 2018), we sought to document the demographic profile of our TAs and sense of belonging, by investigating: 1. What are the demographic profiles of students recruited into TA roles and are they representative of the broader student population? 2. How do students’ demographic characteristics influence their sense of belonging at the start of their TA roles? 2 METHODOLOGY 2.1 Data Collection To address the RQs we collected data during the induction pedagogical training for ~ 400 teaching assistants recruited for the large first year courses (i.e. calculus, physics, computing) in fall 2023. Paper survey was selected as an efficient means to collect anonymous at source information from a large number of people that typically has a high response rate when administered in person. Open-ended questions about experiences students identified as impacting their sense of belonging provided qualitative, descriptive information about students’ experiences. While the nature of our research question involved quite detailed demographic information, we grouped identity categories together in the survey options to improve anonymity and also to produce a tractable dataset to facilitate statistical queries. For instance, we provided only 3 response options for the question « Before coming to EPFL, where did you study? » of CH, EU/UK and other. As there is ambiguity in the way belonging in higher education is defined (Dias-Broens et al., 2024) and experienced (Lee et al., 2024) we used multiple statements to capture TAs’ sense of belonging, including options about feeling, being a member, or part of a community without ascribing a hierarchical structure between them. This study was approved by the institutional research ethics committee (HREC 052-2023). 2.2 Data Analysis Survey data was imported into SPSS for quantitative analysis (version 28.0.0.0) and MAXQDA 2020 (VERBI Software, 2020) for qualitative analysis. We used a grounded approach for the qualitative data coding, coding each response without predefined codes and allowing themes and categories to emerge directly from the participants’ responses. This inductive method ensured that the findings were reflective of the participants' perspectives and experiences. Quantitative analysis employed statistical tests including Chi squared and Fisher’s Exact test to determine significance differences between groups. Institutional data was used to make statistical comparisons about the demographics (EPFL, 2025). 3 RESULTS 3.1 Demographics and representativity of TAs relative to the general EPFL student population Of the 314 responses collected, we eliminated data from people who had previously been TAs (64 people) and those who did not answer this item (6). Our analysis was conducted with the remaining 244 students who had been recruited into their first TA roles for physics (67 TAs), computing (45), maths (72) or other courses (60). We investigated the demographic profile of our respondents in comparison to the overall bachelor and master student body in several aspects including gender, migration background and first-generation status. Due to the nature of the institutional data, each of these comparisons were made independently and therefore do not address issues of intersectionality. Academic affiliation & progression: Most of the TAs were in their second year of studies (71%). We also had TAs in their third year of the bachelor programme (11%) and some pursuing a masters degree (16%). TAs were distributed across the five engineering schools present in the institution, with most of them coming from mechanical engineering and computer sciences. This distribution is not representative of the institutional population. Most TAs were recruited to accompany students from their own academic department (59%). Gender: Our sample is 70% male, which is representative of the EPFL Bachelor/Master student population of 65% male in 2023 (Chi square, p >0 .1). To ensure adequate anonymity yet allow TAs to accurately report their gender, we provided them with a third option “other” which was chosen by 2 TAs (0.8%). While the percent of male TAs varies significantly between the academic departments, our Chi square analysis found our study population remains representative even at this level. Sexuality: TAs in our study population mostly reported heterosexual orientation with 5.1% homosexual and 8.5% choosing ‘other’ designation. While we do not have comparative institutional data, comparison with the general population indicates that our sample is representative (Chi sq., p>.6) (Ipsos, 2021). Migration background & First-generation: EPFL is an international institution, with >130 nationalities represented on campus. Our study population was identical to the institutional student body in that 50% of our respondents arrived from EU or UK institutions. However, domestic students were underrepresented, and students from non-EU/UK countries were overrepresented (Table 1). This may be influenced by the relative financial burden of the cost of life in Switzerland for international students compared to domestic students. Our study population comprised 42 first-generation TAs (18%), which is significantly lower (Chi sq. (2, 244) = 278.8, p <.001) than the 32% first-generation students in the overall institutional population. Table 1. Distribution of respondents’ country of prior education as compared to institutional statistics. Switzerland EU/UK Other TAs 56 (25%) 114 (50%) 58 (25%) Institution 4545 (43%) 5244 (50%) 672 (6%) The overall result of our demographic analysis is that TAs for first year courses are representative of the student body across most aspects, except that first-generation students and domestic students are under-represented relative to their presence at the university. 3.2 Belonging and influence of demographic characteristics The TAs responded to three closed/quantitative items about their sense of belonging at the institution, in their academic department, and in their discipline. Almost all TAs report feeling a sense of belonging at the institution with only 3% (n = 6) reporting that they feel like an outsider. TAs reported an identical high level of belonging in their disciplines to that of the institution, however 7% (n = 16) feel like outsiders in their academic departments (Figure 1). This difference is significant (Fisher’s exact test p < .001) compared to disciplinary and institutional belonging, with a moderate effect size (Cramer’s V .368 and .345, p < .001, respectively). Figure 1. TAs' experience of belonging at the level of their academic department and in their discipline In a condition where there is little variation in a sense of belonging expressed by participants, it is unlikely to uncover relevant differences related to TAs’ demographic profiles. We found only 2 such factors, related to their migration background and the alignment of their own academic discipline with the students for whom they were TAs. TAs’ migration background was relevant to their sense of disciplinary belonging, but not their belonging at the institutional or department level, with people who had at least one parent born outside of EU/UK/CH feeling less belonging in their discipline (Fisher’s Exact test 11.364, p = .003), with moderate effect size (Cramer’s V 0.232, p = .003). Being a TA assigned to course with students from same discipline correlated with a moderate increase in sense of belonging at EPFL (Fisher’s Exact test 16.082, df 3, p<.001, Cramer’s V .283, p< .001) but did not affect TAs’ sense of belonging at the department or discipline. In our study population, we found no significant differences in sense of belonging (Fisher’s exact tests p > .013) based on TAs gender, sexual orientation, academic department, year of study, first generation status, location of previous studies, and location of parents’ birth. 3.3 Experiences that contribute or detract from TAs’ sense of belonging Respondents were asked to give two specific examples of experiences that increased and decreased their sense of belonging, indicating if this acted at the level of the institution, department, discipline or other. Experiences that increased belonging were contributed by 191 TAs and detrimental experiences by 130 TAs (Table 2), with academic success and social interactions as the major themes. Aspects of social belonging including participation in social activities (57), joining associations (35) and engaging in pro-social activities like helping others (51) dominated positive experiences. Academic success, recognition and interest were often cited (55 times) as experiences that increased belonging while opposing experiences of academic failure and lack of interest (35) were common detrimental examples. Experiences that were detrimental for social belonging included discriminatory behaviours from others (27) and isolation (27). Institutional and departmental belonging featured most often in students’ examples (Figure 2). Almost half students indicated that their example, positive or detrimental, affected more than one sphere of belonging. Table 2. Specific examples of experiences that increased or decreased students’ sense of belonging Positive experiences Participating in social activities “Section apéros with all my colleagues” Academic performance and sense of achievement “The CM project made me love my section and showed me that I made a good choice.” “In the first weeks of my [first semester], I solved an analysis exercise by using a different method than the 'expected' one. My assistant praised me for that.” Participating in altruistic activities “When the second year started, there were many new students entering the telegram chat of EPFL and I felt good when I was able to answer their questions.” “helping some others and getting help” Positive one-on-one interactions “Getting to talk with professors, feeling as they know me, and I am not just another number, another student just waiting to be expelled.” “Most of the people I talk to, whether it be classmates, TAs, teachers, they are all very friendly and try to help.” Joining groups, associations “Taking part in an association … made me really feel like I was part of the EPFL community & section” Common experiences, interests “During exam period, I see everyone working hard to pass and [unreadable] I find myself talking to strangers about how tough it is and how we all must go through it. May seem simple, but it for sure increased my sense of belonging to the [institution] community.” Detrimental experiences Poor academic performance and not liking the subjects “Not passing the first year at the first try.” “Having a hard time understanding physics and maths classes during my first year and not feeling comfortable enough to ask TAs for help” No examples “I don't think of an experience that decreased my sense of belonging.” Unwanted discriminatory interactions “Elitist behaviours. Students, professors or assistants who bring down other students.” “not ALL people are open minded, …: still hear homophobic comments while on campus.” Isolation “Having to follow classes virtually due to problems finding housing near campus during my BA1 semester.” “I felt like I was 'avoided' because I had an accent while talking in French + because I stayed at 20km from EPFL people didn't really want to bond.” Figure 2. Sphere of belonging affected by TAs’ examples of positive and detrimental belonging 4 DISCUSSION AND CONCLUSIONS TAs for first year courses are representative of the institutional student population on most demographic characteristics, which should have a positive impact on their capacity as near-peer role models. However, domestic students and first-generation students appear less likely to be TAs. That first generation TAs are underrepresented versus the general student population by nearly half is especially important because the first year of studies at EPFL is a highly selective year with a high failure rate. This means that we are not benefitting from the capacity of firstgeneration TAs to better support peers who share this characteristic (Atkins et al., 2020; Dennehy & Dasgupta, 2017; Estrada et al., 2022; Muir, 2018) and therefore missing an important opportunity to support the success of these students identified as being at increased risk of attrition. The paucity of domestic student TAs may be a product of the relative financial burden of the cost of living in Switzerland leading to more international students seeking part-time employment during their studies. TAs in the large, first year courses have a high sense of belonging before starting their TA roles. Due to the relatively homogeneous level of belonging expressed by participants, we found little correlation between aspects of TAs’ identity and the quantitative belonging items. Another study that explored engineering students’ sense of belonging found no differences based on gender (Becerra-Cid et al., 2023). Aspects related to identity did feature in some of the specific experiences related by students, such as discriminatory interactions that decreased belonging. Social inclusion and academic success feature as major themes in the experiences that students cited as shaping their sense of belonging. Previous studies have similarly shown that discussing subject matter with peers outside the class, engaging in social organisations, and positive interactions with faculty (both during and outside class) increase sense of belonging (Christe, 2013; Hurtado & Carter, 1997; Strayhorn, 2018) and therefore retention, especially in minoritized students (Chang et al., 2014; Litzler & Samuelson, 2013). As these characteristics are often positively correlated to being recruited as TAs, it is likely that our sample of TAs have a higher sense of belonging than the general student population. Social exclusion is a major detrimental factor for sense of belonging and can arise from difficulty participating in social activities as a result of being a commuter student (Newbold et al., 2011) or the lack of financial resources (McClure & and Ryder, 2018). This is especially relevant in our context as Switzerland has a high cost-ofliving and there is a general shortage of housing, particularly for students. We have not produced a satisfactory answer for our second RQ. While we did not find any differences in sense of belonging based on TAs’ identity, this may or may not arise from the lack of sufficiently nuanced data. While repeating a similar study with a more sensitive instrument (multiple-question scale constructed from 8-point Likert items) would be one option, our data is sufficient to indicate that sense of belonging is likely not a barrier for students already recruited into TA roles. Rather, it would be more relevant to assess how students’ identity influences the sense of belonging of students who apply for TA roles and those who do not. Generating richer qualitative data about the interaction between sense of belonging and aspects of students’ identity with specific prompts is another potential approach. Another interesting avenue would be to document longitudinal changes in sense of belonging across a TA experience, as this likely creates opportunities for academic recognition and social integration. Engineering schools have a moral imperative to support the academic success of all students, and therefore to consider the implications of how TAs are selected and trained. We have identified that domestic and first-generation students are underrepresented as TAs, highlighting an avenue to improve support for an at-risk student population. While TAs expressed a healthy sense of belonging, we found instances of elitist and discriminatory behaviour decreasing students’ sense of belonging. We encourage institutions to support TAs’ expectation that inclusion is part of their job (de Lima et al., 2023, 2024) by ensuring TAs are aware of the diversity in the student population and therefore able to accompany their peers more effectively.