Authors: Colin Mack (Course Convenor), Alissa Hultman (PAL facilitator), Disha Borse (PAL facilitator), Natalia Krysztofiak (PAL Trainer), Samayla Mahjabin Koishy (PAL Trainer), Robert Imbur (PAL Lead)
Why our quantitative methods course needed a peer space
Quantitative methods courses often occupy an uneasy place in social science programmes. Students may recognise their importance, but still approach them with anxiety, uneven prior experience, and uncertainty about what it means to “think quantitatively”. On our postgraduate course Quantitative Data Analysis (SPS5033), this was especially visible in the gap between understanding concepts in lectures and feeling confident enough to apply them independently in data analysis tasks. Peer Assisted Learning (PAL) was introduced as one way of approaching this challenge: not to replace lectures, tutorials or staff support, but to create an additional, student-led space where learners could revisit difficult course material, ask questions informally, and hear from peers working through similar problems. In practice, PAL took the form of 13 voluntary drop-in sessions across the term, running on Wednesdays and Fridays, led by trained student PAL leaders and focused on students’ quantitative project work in R, the statistical programming language used on the course.
The aim was threefold: to support confidence in quantitative methods; to provide a less intimidating space for practice and discussion; and to build a stronger learning community around a subject that students can otherwise experience as isolating. This reflects wider work on Peer Assisted Learning, which positions peer learning as a collaborative, non-remedial form of academic support (Dawson et al., 2014; Garcia-Melgar et al., 2021; Bhat et al., 2022; Dekker et al., 2025). It also connects with research on statistics anxiety (Onwuegbuzie and Wilson, 2003; Macher et al., 2012), which suggests that students’ emotional relationship with quantitative work can shape engagement, confidence and willingness to persist with statistical learning (e.g. Trassi et al., 2022; Hernández de la Hera et al., 2023; March et al., 2025; Muerza et al., 2026).
What PAL made possible on the course
The reflections below draw on written comments from PAL leaders, especially Alissa and Disha, who facilitated sessions during the course. They are therefore not presented as formal evaluation data, but as practitioner reflections on what PAL made possible and where the model could be strengthened.
One of PAL’s main strengths was that it created a lower-stakes environment for working through difficulty. Students came with R problems, coding questions, decisions about variables, and uncertainty about model specification. Rather than experiencing these problems individually, they encountered them collectively. As Alissa reflected, it was valuable for students’ morale to see others working through similar issues. This matters in quantitative methods because students can easily interpret confusion as evidence that they are “not good at statistics.” PAL helped normalise difficulty as part of the learning process.
A second strength was the emphasis on articulation. Disha noted that one of the most useful practices was reading through code aloud and explaining what it was doing and why. This moved students beyond simply trying to make code run. It encouraged them to explain their choices, compare approaches, and think more carefully about the relationship between coding decisions and analytical reasoning. PAL also revealed differences in how R was taught across tutorial groups. This sometimes created confusion, but it also exposed students to alternative routes through the same problem. PAL supported the development of quantitative judgement in this sense: that quantitative analysis is not always about finding one correct command or approach, but about making defensible choices.
The sessions also generated student-led resources. When attendees said they were struggling to break down the final assignment requirements, the PAL leaders worked through the course handbook, example answers and intended learning outcomes to produce a chronological checklist. This became a shared resource to support students from dataset selection through to diagnostic checks and final interpretation. Importantly, this was not simply a staff-designed intervention delivered by peers; it emerged from the facilitators’ understanding of what students were finding difficult in real time.
The sessions also showed the reciprocal value of PAL for facilitators. Both facilitators described how supporting others strengthened their own understanding. Because each student project was different, facilitators encountered a wider range of datasets, coding decisions and modelling problems than they would have through their own work alone. PAL leadership also supported their own discipline and time management, as preparing for sessions encouraged them to stay close to the course material throughout the term.
At the same time, the reflections point to some challenges. The first was attendance. While students who attended appeared to benefit, participation was fairly low, uneven and increased near the summative deadline. Students who might benefit most may be least likely to attend, especially if they feel anxious, busy or unsure what PAL is for. A second challenge was facilitator confidence, especially in the initial sessions. Although PAL is peer-led rather than subject-specialist tutoring, Alissa noted that it would be useful for at least one facilitator in each session to have a stronger background in R. This suggests that PAL should remain peer-led, but not unsupported.
Keeping PAL peer-led, but better embedded
These reflections suggest that PAL’s value lay not simply in offering extra support, but in creating a different kind of learning space around quantitative methods. It gave students access to peer explanations, reassurance, and practical strategies from those navigating similar issues on the same course. In doing so, it helped reframe quantitative methods as a process students learn through practice, conversation and repeated attempts, rather than something they either “get” or “don’t get”.
In future iterations, we would strengthen the structure around PAL while preserving its peer-led character. This could include clearer preparatory frameworks for facilitators, more regular check-ins, and involvement from former facilitators or PAL leaders with stronger confidence in R. We would also communicate PAL more explicitly as proactive learning support for all students, including as a way of building regular study habits rather than turning to support only near the deadline. A clearer week-by-week structure, aligned with lecture topics, assessment milestones and student-requested themes, could also help embed PAL more visibly into the course. Rather than presenting PAL as a separate support offer, we would consider integrating it more explicitly into lecture discussions, assignment preparation and revision points, so that students encounter it as a normal part of practical learning on the course. The student-generated checklist could also be refined and carried forward as part of this model. Overall, this would require closer coordination between PAL leaders and teaching staff, so that sessions complement formal teaching while retaining their distinctive peer-led character.
References
Bhat, N., Gurung, S., Gupta, M., Dhungana, N. and Thapa, R.K. (2022) ‘Enhancing collaborative learning through peer-assisted learning’, Journal of Physiological Society of Nepal, 3(1), pp. 4–9. doi: 10.3126/jpsn.v3i1.57762.
Dawson, P., van der Meer, J., Skalicky, J. and Cowley, K. (2014) ‘On the effectiveness of supplemental instruction: A systematic review of supplemental instruction and peer-assisted study sessions literature between 2001 and 2010’, Review of Educational Research, 84(4), pp. 609–639. doi: 10.3102/0034654314540007.
Dekker, I., Koerhuis-Pasanisi, M.-J. and Koek, M. (2025) ‘When and how do peers stimulate engaging in desirable difficulties: Student perspectives on the effectiveness of supplemental instruction’, Active Learning in Higher Education, 26(2), pp. 365–380. doi: 10.1177/14697874241249130.
Garcia-Melgar, A., East, J. and Meyers, N. (2021) ‘Peer assisted academic support: A comparison of mentors’ and mentees’ experiences of a drop-in programme’, Journal of Further and Higher Education, 45(9), pp. 1163–1176. doi: 10.1080/0309877X.2020.1851665.
Hernández de la Hera, J.M., Morales-Rodríguez, F.M., Rodríguez-Gobiet, J.P. and Martínez-Ramón, J.P. (2023) ‘Attitudes toward mathematics/statistics, anxiety, self-efficacy and academic performance: An artificial neural network’, Frontiers in Psychology, 14, article 1214892. doi: 10.3389/fpsyg.2023.1214892.
Macher, D., Paechter, M., Papousek, I. and Ruggeri, K. (2012) ‘Statistics anxiety, trait anxiety, learning behavior, and academic performance’, European Journal of Psychology of Education, 27(4), pp. 483–498. doi: 10.1007/s10212-011-0090-5.
March, J.J., Hamilton, D., McCormack, D., Brisco, R. and Grech, A. (2025) ‘A network analysis of statistics anxiety symptoms and their antecedents in UK higher education students’, Annals of the New York Academy of Sciences, 1547(1), pp. 220–232. doi: 10.1111/nyas.15350.
Muerza, V., Gargallo, P., Salvador, M. and Turón, A. (2026) ‘Positive emotions, problem-based learning and the development of sustainable competencies in higher education statistics’, Sustainability, 18(8), article 3728. doi: 10.3390/su18083728.
Onwuegbuzie, A.J. and Wilson, V.A. (2003) ‘Statistics anxiety: Nature, etiology, antecedents, effects, and treatments — A comprehensive review of the literature’, Teaching in Higher Education, 8(2), pp. 195–209. doi: 10.1080/1356251032000052447.
Trassi, A.P., Leonard, S.J., Rodrigues, L.D., Rodas, J.A. and Santos, F.H. (2022) ‘Mediating factors of statistics anxiety in university students: A systematic review and meta-analysis’, Annals of the New York Academy of Sciences, 1512(1), pp. 76–97. doi: 10.1111/nyas.14746.