Medical Students’ Participation in Social Studying and Learning during COVID-19

Abstract

COVID-19 transformed the medical school learning environment. How social studying and learning (SSL) may have changed was considered worthy of exploration. This study describes the effect of the learning environment transformation on how SSL is conducted by medical students and the reasoning behind any changes. A post-positivist stance was adopted with a mixed method convergent-parallel approach. An online survey explored the participation rate and nature of SSL and how these related to literature-identified underpinning factors that influence participation in SSL. A follow-up interview explored the survey responses. A total of 87 survey responses were collected. Following exclusion of incomplete responses, 57 responses were analysed, and nine follow-up interviews were conducted. Cross-tabulation and logistic regression were conducted to analyse the quantitative data and thematic analysis was conducted to analyse the qualitative data. No significant difference was observed in SSL participation rate during COVID-19. Students transitioned from in-person to Zoom meetings for sensibility, ease and convenience. Students continued participating in SSL for motivation and began participating for accountability, focus and replacement of lost social interaction. The same content was studied using a wider range of activities. Significant decreases were observed in students’ perception of the effect of a range of factors. The observed changes to SSL during online learning suggest that SSL is integral for many medical students and will be adapted as required. Given that medical school will always involve some periods of online learning, understanding of the nature of and driving factors behind how medical students engage in SSL during online learning may enable medical educators to support all aspects of student learning.

1. Introduction

Due to the vast breadth of information in the field of medicine, progressing through medical school requires medical students to learn large amounts of information in short amounts of time. A significant component of medical student learning is self-directed study, sometimes with peers [1,2], that occurs away from the classroom and the well-researched pedagogical approaches employed within it. How medical students approach self-directed study has been investigated with limited scope [3,4,5,6,7], mainly focussing on the effect that different learning strategies and techniques have on performance in high stakes examinations. However, there is limited research into the nature and role of self-directed study with peers amongst medical students.

In their groundwork study that explored how medical students study with their peers, Keren et al. [8] (p. 353) defined social studying and learning (SSL) as “any independent, elective, self-directed, and self-organised approach to learning that involves students working with their peers for the purposes of study, learning, or revision”. SSL is different to faculty-organised group work such as problem-based learning and small group tutorials, where the learner develops different skills [9] and may participate differently due to a fear of scrutiny from staff [10] or a lack of personal connection with others in the group [11].

Due to the perceived advantages, many medical students value participation in SSL [1]. These advantages include greater engagement with course material [12,13,14], increased motivation and accountability [15], provision of support and wellbeing [16] and development of communication and teamwork skills [14,17]. Participation in SSL has been shown to not have a direct impact on examination performance [13,18]. This is consistent with the model proposed by Keren et al. [8], in which students who do not want to participate in SSL are not disadvantaged.

Not all medical students participate in SSL [1,19]. Whether a medical student participates in SSL is influenced by a network of intrinsic and extrinsic factors [8]. The impact of a factor varies depending on a student’s circumstances—a factor can encourage participation in SSL for one student but be a barrier for another. Intrinsic factors include a student’s social acumen and confidence, academic ability and needs relative to their peers, self-efficacy and agency, whether they seek social interaction and support, and the nature of previous SSL experiences [8]. Extrinsic factors include a student’s location relative to their peers and university, use of social networks, other time commitments, the nature of the class dynamic and at what point in their studies they seek SSL [8]. Each student’s combination of factors contributes to their decision whether to participate in SSL. A study of medical students in Australia found that students participated in SSL if they had high self-agency due to stress about examinations, if they believed they learned best with other people, if they had received advice from other students and staff, or if they were invited to SSL by another student [11]. Conversely, students did not participate if they lacked self-agency or had difficulty finding a group, if they thought SSL was not useful, if they believed they learned best on their own or if they were overwhelmed or intimidated by others [11].

The COVID-19 pandemic resulted in the temporary closure of universities worldwide. Online learning replaced face-to-face learning [20] and medical courses were no exception [21,22,23,24,25,26,27,28,29]. The effect of online learning on how medical students participated in SSL has had limited exploration [30,31]. Until online learning, medical students who participated in SSL usually met at least once weekly for between one and three hours [13] on the university campus [32]. SSL was usually with friends and involved teaching each other, quizzing each other and making or reviewing topic summaries [13].

Given the health restrictions on gatherings [33], the authors hypothesised that the continuation of SSL in the same way would be challenging, and hence there would be changes to how SSL was conducted. How the factors identified in the literature [8,11] may have changed and impacted on students’ decision making whether to participate in SSL was considered worthy of exploration. This study aims to describe how and why SSL conducted by medical students changed when learning transitioned online owing to the COVID-19 pandemic.

2. Materials and Methods

2.1. Research Design

The phenomenon being studied is the possible effect that the medical school learning environment transformation due to COVID-19 may have had on social studying and learning (SSL). Through their study, Keren et al. [8] suggested several factors that affect SSL participation. The phenomenon is therefore assumed to have a tangible effect on SSL participation and can be directly measured. The paradigm adopted by this study is post-positivist, which recognises that bias is inherent in the research. By adopting this stance, we acknowledge that this study is not purely a “problem-solving” piece of research but is also “problem-setting”. This open-ended, exploratory approach may lead to further questions regarding SSL that the authors have not anticipated, along with the construction of new meanings and knowledge regarding the phenomenon being studied [34].

2.2. Participants

This study was conducted at an Australian medical school in 2020. Participants were enrolled in a graduate-entry medical degree with two years pre-clinical teaching on university campus followed by two years clinical placement at healthcare facilities. Due to COVID-19, pre-clinical students transitioned to online learning by April and remained online until 2021. Clinical students had transitioned to online learning by April and were removed from placement for six weeks. The ethics of this study were approved by our University Human Research Ethics Committee–Protocol 2020/317.

2.3. Research Instruments

This study used a mixed method convergent-parallel approach [35] that included a survey designed to characterise a participant’s involvement in SSL (Appendix A) and an optional follow-up interview to further explore their responses. Since an existing validated tool pertaining to SSL could not be found in the literature, a new survey was designed. The survey comprised 31 questions in two halves that characterised a participant’s involvement in SSL before and during online learning respectively. Each half contained multiple choice and free response questions that characterised participants’ involvement, followed by 17 five-point Likert scale questions that assessed participants’ perception of how factors may have influenced their participation in SSL. The factors were derived from a previous study in which a model of intrinsic and extrinsic factors that influenced participation in SSL was synthesised [8]. The Likert scales were chosen to be five-point to reduce confusion and encourage survey completion [36]. The face validity of the survey was assessed by all three authors to ensure that the questions effectively characterised all aspects of how and why SSL was conducted before and during online learning, and that the questions were simple, clear and non-leading.

A semi-structured interview approach was designed by the first author. The approach started with open-ended questions about the participant’s involvement in SSL both before and during online learning, continued with questions directed at the participant’s individual survey responses and finished with an opportunity to make any general comments. To ensure validity of the interview protocol, a draft of the indicative questions was reviewed by the third author (an experienced interviewer), and a practice interview was conducted.

2.4. Data Collection

All enrolled students (n = 422) were invited to participate in the survey via the university learning management system. Participation in this study was voluntary. The survey used Qualtrics and was open for one month during online learning. A reminder and Facebook post in the private group of each cohort was made after two weeks, totalling three distributions of the survey. Participants who provided their email address at the end of the survey were invited to attend a 20-min interview. The audio-recorded interviews were conducted by the first author using Zoom, transcribed verbatim using Otter.ai and checked manually. Participants who completed the survey only were numbered S1–S48. Participants who completed both the survey and follow-up interview were numbered P1–P9.

2.5. Data Analysis

Quantitative survey data were analysed using IBM SPSS Statistics 26. Participation in SSL before and during online learning was compared using McNemar’s test. The change in each factor from before to during online learning was assessed using paired sample t-tests. Logistic regression was run to investigate the predictive effect of the surveyed factors on participation in SSL, both before and during online learning. First, the scale responses were dichotomised, with responses of three or below classified as negative and responses of four or above classified as positive. An individual logistic regression was performed for each factor. Any factor that predicted participation with an odds ratio below 0.5 or above 2.0 was included in a combined logistic regression.

All qualitative survey data and interview transcripts were analysed by the first author. Thematic analysis was conducted, guided by the Association of Medical Education in Europe Guide to Qualitative Analysis [37]. Due to the small sample size, all data were analysed without using computer software [38]. Initial codes were generated for each survey and interview response. Initial codes were then visually clustered into themes using a thematic map, noting frequency where many of the same code were observed. Each theme cluster was named and reported as part of the synthesis of results.

3. Results

3.1. Response Rates

A total of 87 responses were recorded from 422 invited students (21%). Of these, 30 responses were excluded because less than half of the survey was completed, leaving 57 responses (14%). In total, 38 responses were from pre-clinical students in the first two years of the degree and 19 responses were from clinical students in the last two years of the degree. Twelve participants opted in for an interview. Nine participants were interviewed, all of whom were pre-clinical students, while three did not respond to the invite.

3.2. Participation in SSL

There was no difference in the overall SSL participation rate before and during online learning (63% vs. 61%, p = 1.000).

3.3. Medium and Location of SSL

Before online learning, all but one student who participated in SSL did so in person as it was convenient (S21) and where productivity was highest (S13). SSL occurred more commonly in tutorial rooms and university residence study rooms and less commonly at home, the common room and the library. The tutorial and study rooms were advantageous as they had a whiteboard (S7) and big screen (S12).

During online learning, students participated in SSL online by necessity (S9). Students thought Zoom was sensible (S5, S33), easy to use (S6, S35) and convenient (S1, S7, S21). Students occasionally used other media including shared documents and Facebook groups.

3.4. Frequency, Length and Organisation of SSL

Before online learning, most students met once weekly for one to three hours. Some met at the same time each week while others met sporadically. Three students met fortnightly (S4, S26, S30) with one student increasing frequency before an examination (S4). Meetings were commonly organised using Facebook Messenger (P8).

During online learning, the length and organisation of SSL remained similar. Some students met more frequently since Zoom was easy (S6) and helped maintain routine (S1). Meeting once weekly was most common as it facilitated revision of difficult concepts (S4), fitted around other commitments (S5, S11, S20) and aided tutorial preparation (S16).

Some students had difficulty organising SSL because other students returned to their family homes:

“They had other commitments that made organising a time when everyone can meet a little bit more difficult logistically.”

—P8

Conversely, one student reported that the ease of organising a Zoom meeting helped with regularity, meaning that they got more out of SSL (P4).

3.5. Composition of SSL Group

Before online learning, no one group size was more common than another. Most groups comprised friends, peers or both. Some students placed importance on social aspects, especially in first year:

“It was important to meet some people to help you get through social wise, but also to rely on academic wise.”

—P6

Other students placed importance on studying with others from a similar background:

“I haven’t been in formal university studies for 10 years. And so, it was nice to have that sort of peer support and people come at it from a similar level to me.”

—P3

During online learning, the group size also varied. One student preferred a smaller group due to the limitations of online communication (P4). Students pursued SSL for social interaction:

“But partly was just trying to navigate the lack of contact with people who are studying the same material.”

—P7

“I moved back home during the online learning period. And because I live rurally… I felt like I was lacking a lot of support.”

—P5

3.6. Subjects and Activity Type during SSL

Before online learning, activities included individual study in a group setting, general discussion, “quizzing” each other, making summaries and practising clinical skills. These activities covered difficult subjects. Only one student completed prescribed coursework during SSL (P8).

“We’d discuss the topics, we’d test each other on things, we’d ask each other rapid round questions. And we’d, if someone was struggling with the topic, use the opportunity to teach them a little bit.”

—P4

Most students reported they covered “everything in the course”. Several students placed importance on preparing for SSL:

“Important to give other people a bit of a heads up, so we can all prepare a bit, because I hate getting stuck on a topic.”

—P6

During online learning, students completed a larger range of activities. Activities completed before online learning continued, except clinical skills, which halved in frequency. New activities included going through clinical cases by annotating an electronic textbook (P4) and working through questions written by previous students (P5). One student found “quizzing” each other was the easiest activity using Zoom (S13).

One student tried many different activities (P7). Initially, they studied independently over Zoom but that “fizzled out”. They also joined two “learning co-operatives” and met in person again once health restrictions eased:

“Sign-up system where people write (flash) cards, upload them to the cloud so that we can all access.”

—P7

Students still studied “everything”. One student elaborated that they focussed on “anything that was relevant at the time” (S2).

3.7. Reasons for Participating in SSL

Before online learning, students most commonly participated in SSL for motivation, including for topics they found hard (P8). Another student reported that participating helped them focus:

“If I hadn’t had other people around me, I probably would have fallen into a trap of going on Facebook…”

—P7

During online learning, students participated in SSL for motivation, accountability and focus. Students found that the presence of others helped them ensure thorough work (P6) and maintain attention:

“But having someone on Zoom on the other end, I feel has contributed to my focus and attention.”

—P2

3.8. Factors Suggested to Influence Participation in SSL

Significant decreases were observed in the perceived effect of three intrinsic factors on participants’ involvement in SSL during online learning (

Table 1. Mean ratings of the effect that intrinsic and extrinsic factors had on participation in social studying and learning, both before and during online learning.

Factor	Mean Rating (Out of 5) ± Standard Deviation		p Value †
	Before Online Learning	During Online Learning
Intrinsic factors
Social acumen and confidence	3.57 ± 0.683	3.09 ± 0.880	0.000 *
Academic ability relative to your peers	3.19 ± 0.770	3.15 ± 0.780	0.710
Ability to manage your own learning	3.62 ± 0.822	3.28 ± 1.057	0.044 *
Social interaction and support network	4.00 ± 0.860	3.09 ± 1.139	0.000 *
Previous experiences of social studying and learning	3.43 ± 0.807	3.43 ± 0.807	1.000
Extrinsic factors
Location relative to your peers	3.43 ± 1.223	2.52 ± 1.130	0.000 *
Location relative to [University] campus	3.59 ± 1.343	2.50 ± 0.960	0.000 *
Location relative to the [City] Hospital	3.26 ± 1.104	2.70 ± 1.030	0.000 *
Usage of technology generally	3.70 ± 0.940	3.89 ± 1.178	0.192
Usage of social media	3.69 ± 0.925	3.67 ± 1.128	0.868
Other time commitments	2.93 ± 1.020	2.83 ± 1.018	0.481
Attitudes of peers	3.62 ± 0.834	3.56 ± 0.943	0.596
Attitudes of staff	3.36 ± 0.773	3.09 ± 1.041	0.044 *
Advice from peers	3.82 ± 0.684	3.58 ± 0.723	0.040 *
Advice from staff	3.58 ± 0.723	3.29 ± 0.843	0.014 *
Advice from friends	3.56 ± 0.785	3.47 ± 0.726	0.377
Advice from family	3.40 ± 0.751	3.40 ± 0.780	1.000

* denotes significance of p < 0.05. † Significance was determined using paired sample t-tests.

). The decreases in social acumen and confidence (p < 0.001) and in social interaction and support network (p < 0.001) were largely attributed to having no in-person contact with peers (P1, P5, P9). The decrease in one’s ability to manage their own learning (p = 0.044) was attributed to several reasons, including having a less rigid timetable (P1), increased presence of distractions such as a smartphone (P7) and asynchronous learning:

“I think that I found the hardest thing with online learning was that everybody’s learning level was no longer at a similar point… and so it meant there wasn’t a commonality about where we’re up to and what we’re doing.”

—P3

Significant decreases were observed in the perceived effect of six extrinsic factors on participants’ involvement in SSL during online learning (Table 1). Decreases were observed in all location-related factors—location relative to one’s peers (p < 0.001), university campus (p < 0.001) and the hospital (p < 0.001)—which was because participating in SSL using Zoom did not require travel (P7, P8). One student thought that Zoom may have encouraged their group members to participate:

“Some people may have found it easier to log on to a Zoom meeting than to commute to campus and attend in person.”

—P8

However, one student thought that in-person learning was more conducive to group formation as all students were physically together (P9).

Decreases were observed in the perceived attitudes of staff towards SSL (p = 0.044), advice from peers (p = 0.040) and advice from staff (p = 0.014). One student described this as a reflection of the spectrum of staff attitudes towards online learning (P9).

3.9. Predictive Effect of Factors on Participation

Based on the individual logistic regression odds ratios (

Table 2. Results of the individual logistic regressions for each factor before and during online learning to determine whether each had a predictive effect on participation in SSL.

Name of Factor	Before Online Learning			During Online Learning
	Significance	Odds Ratio	95% Confidence Interval	Significance	Odds Ratio	95% Confidence Interval
Intrinsic factors
Social acumen and confidence	0.324	1.729	0.583–5.128	0.799	1.177	0.336–4.118
Academic ability relative to your peers	0.361	0.571	0.172–1.898	0.365	0.552	0.153–1.996
Ability to manage your own learning	0.185	2.095	0.702–6.254	0.701	0.802	0.261–2.469
Social interaction and support network	0.697	1.295	0.353–4.750	0.618	0.742	0.229–2.401
Previous experiences of social studying and learning	0.504	1.454	0.485–4.356	0.614	0.754	0.251–2.263
Extrinsic factors
Location relative to your peers	0.707	1.229	0.418–3.612	0.292	0.422	0.085–2.099
Location relative to [University] campus	0.816	1.136	0.386–3.345	0.999	0.000	-
Location relative to the [City] Hospital	0.106	0.400	0.132–1.216	0.075	0.206	0.036–1.175
Usage of technology generally	0.170	0.447	0.141–1.413	0.487	0.679	0.227–2.026
Usage of social media	0.791	0.862	0.286–2.594	0.572	1.364	0.464–4.008
Other time commitments	0.300	0.542	0.170–1.727	0.695	0.750	0.178–3.162
Attitudes of peers	0.357	1.667	0.563–4.937	0.752	0.842	0.289–2.450
Attitudes of staff	0.106	0.400	0.132–1.216	0.799	1.177	0.336–4.118
Advice from peers	1.000	1.000	0.319–3.132	0.685	0.800	0.272–2.350
Advice from staff	0.662	0.786	0.266–2.320	0.897	0.923	0.276–3.084
Advice from friends	0.707	0.813	0.277–2.390	0.701	0.802	0.261–2.469
Advice from family	1.000	1.000	0.319–3.132	0.915	1.067	0.324–3.509

), four factors—ability to manage one’s own learning, location relative to the hospital, usage of technology generally and attitudes of staff (odds ratios 2.095, 0.400, 0.447 and 0.400 respectively)—were included in the logistic regression for before online learning, yielding an insignificant model (χ² = 8.855, p = 0.263). Three factors—location relative to peers, location relative to university campus and location relative to the hospital (odds ratios 0.422, 0.000 and 0.206 respectively)—were included in the logistic regression for during online learning, yielding an insignificant model (χ² = 7.129, p = 0.129). None of the factors significantly contributed to the models.

3.10. Other Barriers to Participation in SSL

Our study revealed some newly reported factors that may have influenced SSL participation during online learning. Some students encountered barriers using Zoom, finding it less personable (P3), difficult to use due to a poor internet connection (P8) and logistically difficult:

“The flow of it didn’t feel natural or as normal as in-person study groups.”

—P7

“I think doing [SSL] on Zoom eliminates a lot of the mediums that are effective.”

—P9

Several other newly reported barriers were encountered by students before online learning. One student reported that they were still settling in and did not know where to start (P9). Another student from a non-science academic background was still working out their study routine:

“I hadn’t developed any study groups at that stage, just because I didn’t know the people that well yet and hadn’t gotten a proper routine.”

—P2

Another student regretted not participating, attributing it to being absent for the first two weeks of the course (P5). However, these three barriers were only encountered by first-year students.

4. Discussion

We aimed to describe how and why social studying and learning (SSL) changed due to the transformation of the learning environment owing to the COVID-19 pandemic. Several changes to how and why SSL was conducted were observed, yet there was no significant difference in the overall SSL participation rate. In what follows, we discuss the changes found and potential reasons why the overall SSL participation rate did not change, and conclude that many medical students consider SSL integral and will adapt it as required.

SSL transitioned from in-person to online using tools such as Zoom as it was easy, convenient and sensible. The latter two reasons reflect previously reported advantages of online learning during the pandemic, including flexibility with timing [21,24,25] and maintaining limited social contact [28]. Conversely, the ease of participating online in our study differs from other studies, which report difficulty due to limited non-verbal communication [25] and bad internet connections [21,24,25,28]. While the students in our study found that SSL was less personable and logistically difficult at times, this was mostly overcome by the adaptation of activities to an online format and did not lead to non-participation. Only one student reported that a bad internet connection was a barrier to their participation.

Students rated their location relative to their peers, the university campus and the primary teaching hospital as less important to their participation. Given the online format, location-related influences such as travel time, method and cost were no longer important.

Many students reported more competing time commitments—particularly those who moved back to their family home. This is consistent with medical students in the United Kingdom who reported that family distractions were the largest disadvantage of online learning [24]. Conversely, other students either desired to or did meet more often. While this could be explained by the ease of meeting online as previously discussed, it may also be explained as a way of counteracting the observed decrease in two intrinsic factors—social acumen and confidence, and social interaction and support network. It is known that medical students pursue SSL for social and emotional support [1,13,16] and meeting more often may have been a substitute for the lost social interaction.

Although not reported in other studies, students covered all aspects of the course in SSL, which did not change during online learning. The aspect of clinical skills was the only exception. Similar difficulties with clinical skills were observed among medical students in Saudi Arabia, which was also linked to a lack of access to the clinical environment [25].

A greater range of activities were attempted in SSL during online learning. “Why” remains elusive but may be related to the uncertainty of a completely novel way of participating in SSL. As students determine what they prefer, they are likely to try a larger number of activities. However, one student suggested that their change in SSL activities may have been the normal evolution of their SSL technique rather than the influence of online learning. Given that most of the interview participants were pre-clinical students who had recently started medical school, it is possible that they were still determining what they preferred and so changes were unrelated to online learning.

Reasons for engaging in SSL during online learning remained similar to before online learning, with the exception of increased importance placed on accountability and focus. Since medical students normally seek SSL for motivation and accountability [1], this change may represent increased difficulty maintaining focus on coursework generally during online learning. This difficulty is consistent with the lack of motivation and concentration encountered by UK medical students during online learning [24]. In our study, one student stated that the comparison with other students gained through SSL was most important for their motivation during online learning, a recognised phenomenon among medical students [15].

One major barrier to participation in SSL during online learning was that some students believed that other students were no longer at the same point in their learning, largely due to the increased flexibility of timetabling and resultant ability to prioritise some learning activities over others. Flexibility was noted as one of the major advantages of online learning among medical students in the UK [24]. However, our data suggest that while flexibility may be advantageous for coursework, it is disadvantageous for SSL. Students place importance on being at a similar point in their learning to their peers when participating in SSL [8]. Therefore, the ability to prioritise some learning activities over others means that students may no longer perceive themselves to be at the same point in learning as others, discouraging them from participating in SSL.

Limitations and Future Recommendations

This study has limitations that may reduce its generalisability. Firstly, only 57 responses were collected in the survey component of the study. According to the Krejcie and Morgan method [39], a further 145 survey responses would have been required to be representative of our medical school population. There may be several reasons for the low response rate. Firstly, participation in this study was voluntary and had no bearing on degree progression. Additionally, multiple other research surveys were circulated to students during the data collection period, possibly creating survey fatigue [40]. However, we contend that the survey and interview data collected provide an insightful picture of how medical students participate in SSL online, and that an increased sample size can be aimed for in future studies.

Secondly, this study involved only medical students at one medical school in Australia. While all participants experienced the same transition to online learning, their experiences may not be reflective of what happened at other medical schools. Thirdly, follow-up interview participants were recruited from students who opted in (convenience sampling), and all participants who opted in for a follow-up interview were pre-clinical students. Consequently, the perspectives in this study may be more representative of pre-clinical medical students rather than all medical students generally. Therefore, we recommend that future studies investigating SSL involve multiple medical schools and expand on the specific perspectives of either pre-clinical or clinical students.

This study did not consider the relationship between SSL participation and academic performance. Future researchers may find it interesting to investigate whether students who started or stopped participating in SSL due to online learning maintained their previous academic performance relative to students whose participation in SSL did not change. Finally, now that much medical school teaching has returned to face-to-face delivery, it may be interesting to investigate whether the changes to SSL observed in this study (such as use of Zoom and a wider range of activities) have been maintained.

5. Conclusions

Despite the challenges presented by online learning, the lack of change in SSL participation rate and the observed changes to SSL during online learning suggest that SSL is considered integral by many medical students and will be adapted as required. While no significant model that predicted participation was found, it is possible that the other reasons for SSL participation, such as the desire for motivation, accountability and focus, which do not directly form part of the factors studied, played a significant role.

Regardless of the transition of much medical school teaching back to face-to-face delivery as public health restrictions were eased, medical school will always involve some periods of online learning. By way of example, it is compulsory for medical students in Australia to complete a rural placement during medical school [41]. Knowledge of how medical students study outside of formal teaching activities is of great value to medical educators, whose role includes supporting all aspects of student learning. Our study confirms that many medical students will engage in SSL during online learning, and medical educators should aim to support this through the provision of high-quality resources and advice.