1 Introduction

In recent years, the increasing adoption of more digital solutions presents opportunities to collaborate from a distance. During the COVID-19 pandemic, education was moved online for safety purposes, which incidentally highlighted the practical benefits of remote meetings (e.g., flexibility, practical convenience, time-saving). There have been extensive debates and research on whether face-to-face or online conditions are better for education (Means et al., 2013). In actuality, we can enjoy the best of both worlds, but there is a lack of research on how to schedule a mix of face-to-face and online classes.

Collaborative learning is where “two or more people learn or attempt to learn something together” (Dillenbourg, 1999). At universities, collaborative learning can occur in tutorial groups and in group projects and represents a valuable skill to prepare students for their careers. When learning collaboratively, students ask questions, build on each other’s reasoning, and discuss disagreements (Dolmans, 2019), to achieve individual and shared learning goals (Strijbos, 2016). One pedagogical model in which students learn collaboratively is problem-based learning (PBL), where students learn through exploring a problem or case in tutorial groups (Hmelo-Silver, 2004). In PBL tutorials at Maastricht University, 10–12 students learn collaboratively by interacting with each other and building a shared understanding of a case, while guided by a tutor. Students are given a problem, they identify known concepts and link it to prior knowledge, brainstorm about possible solutions to the problem, and identify learning goals. This phase is known as the pre-discussion phase. Based on the learning goals formulated in the pre-discussion phase, the students seek relevant sources of information outside the tutorial. When they reconvene, they present and discuss their findings, known as the post-discussion phase. Students are dependent on each other, and each student is responsible for achieving the group’s learning goals.

When students learn collaboratively, they engage in both cognitive and social activities. These socio-cognitive aspects fit well within the community of inquiry (COI) framework, which has been used extensively to research learning communities in online synchronous forum discussions (Garrison et al., 1999). For this study, we use the COI framework to explore how face-to-face and online tutorials are experienced by students and tutors, and to understand what their preferred schedule of a mix of both formats would be. The COI framework represents a process of creating a deep and meaningful collaborative learning experience through the development of three interdependent elements: cognitive, social, and teaching presence. Cognitive presence refers to the degree to which the students can construct meaning through their group discussions. Cognitive presence aligns with PBL, where cognitive presence plays a role when students explore problems, search for information, integrate new information into their cognitive schema, and apply their knowledge. Social presence refers to the ability to project oneself socially and emotionally, indicated by emotional expression, open communication, and group cohesion. Teaching presence is the facilitation of cognitive and social presences to achieve intended learning outcomes, including the teacher’s guidance of knowledge building and direct instruction (Garrison et al., 1999).

Deciding how to schedule face-to-face and online tutorials appears complex because of the differences in cognitive, social, and teaching presence between both settings. Previous studies have reported that online discussions tend to stay on-task (Jonassen & Kwon, 2001; Lantz, 2001), perhaps promoting a more focused environment for cognitive presence. However, students’ well-being may be negatively affected by the lack of social presence and interactions online (Rasheed et al., 2020). It may also be more effortful for tutors, because they may need to provide more guidance online where there may be more ambiguity (Donnelly, 2013). These arguments led us to consider several possibilities in scheduling tutorials or meetings where collaborative learning takes place. One option is a majority of online meetings to facilitate more on-task, focused discussions, with some face-to-face meetings interspersed to encourage social bonding and relief the tutor’s burden. Another option is to schedule more face-to-face meetings at the start of the course to develop social bonds. Generally, scheduling meetings could be done by having various ratios of face-to-face and online meetings, having back-to-back or interspersed settings, or meeting up face-to-face/online when there is a reason to do so. There are also the extreme ends of the continuum, with purely face-to-face and purely online meetings (Hrastinski, 2019), perhaps being a preference for some.

The aim of this study is to explore the experiences and the preferred schedule of face-to-face and online collaborative learning tutorials, based on the perspective of students and tutors. We use the COI framework to structure our exploration, asking the questions: How did students and tutors experience deep learning, social interactions, and the role of the tutor in face-to-face and online tutorials? And how should we schedule future tutorials according to students and tutors?

2 Material and methods

2.1 Design

This qualitative, cross-sectional study utilized semi-structured interviews. This study was approved by the ethics review committee of Maastricht University's Faculty of Health, Medicine and Life Sciences with approval number FHML-REC/2021/119/Addendum01_22.

2.2 Participants

The study was conducted at a Dutch research university in the south of the Netherlands. The university uses a student-centered approach that focuses on small group learning (approximately 12 students per tutorial group). The student population consists of local and international backgrounds. We used convenience sampling, using course announcements to recruit students and tutors from the Faculty of Health, Medicine and Life Sciences. During spring of 2022, we invited participants who were already enrolled at the university prior to the start of the COVID-19 pandemic, and had extensive experience with both online and face-to-face tutorial meetings. The students were made up of third year Bachelor’s students, while the tutors have a teaching qualification in guiding small group learning.

2.3 Procedure

Participants were invited via internal communications, specifically through departmental emails and course emails. Interested participants were given written information before the scheduled interview. One-on-one interviews were carried out with students and with tutors, conducted by H.Q.C. in a private meeting room to ensure confidentiality and comfort for the participants. Most interviews were executed on-site, because we expected the face-to-face setting to present more communication cues than the online setting. However, three interviews were scheduled online per participant’s request. All interviews were audio-recorded with a dictaphone—the Philips DVT6010 (Speech Processing Solutions GmbH, Vienna).

The interview questions were inspired by the COI framework, emphasizing the importance of cognitive, social and teaching presence in online learning (Garrison et al., 1999). For the reason of using simpler language, cognitive presence was operationalized as deep learning, social presence as social interactions, and teaching presence as the role of the tutor. Because scheduling a combination of face-to-face and online educational meetings is a relatively new idea, we first asked about the different experiences in face-to-face and online tutorials regarding collaborative learning, before asking participants to form their ideal schedule. Overall, the interview questions were prepared collaboratively by the authors who have expertise in educational sciences, with consultation by an external educationalist, and co-created through pilot interviews with two tutor participants to check for the comprehensibility and appropriateness of the questions before the main data collection phase commenced. After briefing the interviewees and obtaining their consent, the following materials were used successively to conduct the interviews:

  1. 1.

    Visual aid with explanations (see Appendix 1) depicting different schedules of face-to-face and online tutorial meetings were shown to illustrate the variety of (non-exhaustive) possibilities to schedule the tutorial meetings.

  2. 2.

    Vignettes (see Appendix 2) were presented one by one that describe the COI concepts of cognitive, social, and teaching presence.

  3. 3.

    Semi-structured questions were used to stimulate discussion (see Appendix 3 for interview guide), including:

    • Main questions

      • How was/were deep learning/social interactions/the role of the tutor different in face-to-face and online tutorials?

    • Follow-up questions

    • Closing question

      • What is your final suggestion of how a 2-month course would look like? And why?

      • Show the fill-in-the-blanks schedule (see Appendix 4).

  4. 4.

    At the end of the interview, a demographics questionnaire was presented, which included the following questions:

    • Year of birth

    • Self-reported language proficiency, as more advanced speakers may find it easier to overcome barriers to communication

    • Nationality, because students and tutors from abroad may prefer one setting over the other

    • Years in higher education, describing how many years the students and tutors have been enrolled or teaching in higher education, respectively

    • Other online education experiences, where having more experience may lead to being more familiar and even comfortable with online settings

2.4 Data analysis

Audio-recordings were transcribed verbatim. When anonymizing the participants’ data, the students and tutors were labelled alphabetically. The interview data was coded and categorized thematically, using a mix of deductive and inductive approaches. Deductively, the COI framework was used to guide the vignettes’ questions. Then, each answer given by the participants – personal experience, opinion, description, and explanation – was given a code by the main researcher H.Q.C.. Each code was also labelled with the participant’s overall preferred schedule and whether they were a student or tutor. Codes of similar content were grouped together to form themes. The themes were identified by all researchers (H.Q.C., D.H.J.M.D., M.G.A.O.E., and H.H.C.M.S.) and discussed iteratively to reach consensus. The themes were then expounded upon by summarizing the codes and consolidating them into a coherent explanation.

2.5 Reflexivity

We acknowledge that the background and expertise of the research team contributed to the design, interpretation, and reporting of this study. All members of the research team are involved in providing primarily face-to-face education, with online adaptations made during the COVID-19 pandemic. All research team members are also involved in educational research. The main researcher and interviewer (H.Q.C.) is a second-year post-doctoral researcher, with prior experience in leading two qualitative studies and after an additional 2-day training with an external educationalist to finetune interview competencies specific to this study. The remaining authors (D.H.J.M.D., M.G.A.O.E., and H.H.C.M.S.) are Professors in education, with extensive experience in guiding small groups and using technology to facilitate small group learning. Our different perspectives combined with the theoretical concepts are expected to enhance the transferability of our findings.

3 Results

3.1 Sample characteristics

Data saturation, when novel themes no longer emerged, was achieved after interviews with 17 students and 13 tutors. Sample characteristics are described in Table 1 to provide readers with the context of the ensuing findings, although the authors did not find evidence that the contextual factors could have influenced the results. A typical bachelor student’s schedule tends to vary, depending on the student year and elective courses chosen. The students tended to have a weekly schedule of two tutorials (two hours each), one lecture (one to two hours), and one practical workshop (one hour) per week. The tutors had 38–40 h per week employment, except for one tutor with a 16-h per week part-time employment.

Table 1 Sample characteristics
Full size table

3.2 Themes

Three main themes emerged from the interviews, as summarized and elaborated in Table 2. These themes describe how face-to-face and online tutorials were experienced, with explanations given on the experiences.

  • Theme 1: Online content discussions tended to lack depth.

Table 2 Themes (how), explanations (why), specific behaviors, and quotes from participants
Full size table

Content discussions in online tutorials were perceived as different from face-to-face tutorials (detailed in Table 2). Students and tutors felt that online discussions had the sole purpose of checking off learning goals, while keeping discussions to a bare minimum. Online discussions were either shorter (fewer exchanges) or longer (more silences). There were various contribution obstacles (listed in Table 3) that obstructed the natural flow of discussions in online meetings. Students and tutors reported that online discussions contained fewer exchanges related to applying knowledge to practical examples, giving supporting explanations, sharing personal anecdotes, discussing disagreements, exploring personal interests, and adding jokes. With fewer exchanges online, students spent less time actively learning about the topic. Students and tutors also found online tutorials to be less enjoyable. Student-A explained: “Online tutorials feel more like a chore to me. It’s something you have to do, to get it over with. Well, of course, the face-to-face (tutorials), you also have to do them. But it’s just much more enjoyable. You might find a nice piece of information the day before and then you’re like, ‘Oh, I can’t wait to talk about this with my tutor and peers.’ See what they think, see if the tutor has any more knowledge about this. And you’re just more open to sharing information and really, actually enjoy discussion”.

Table 3 Contribution obstacles during online tutorials that discouraged students and tutors from speaking up
Full size table

In contrast, students experienced a natural flow of discussion in face-to-face tutorials, encouraging them to use their own words. Students reported being able to make more progress in the discussions by covering more details, while having a better understanding of the discussed concepts. Students and tutors also expected that face-to-face tutorials would improve students’ memory of the discussed content, allow students to assess their learning, and build students’ confidence. When tutorials were conducted face-to-face, there was a sense of togetherness, so students felt a sense of responsibility towards the group, to keep each other active, to stimulate each other in the discussions. Even when academic topics were not personally interesting to the students, having a good group dynamic could help them stay motivated and pay attention in class.

  • Theme 2: Sense of disconnection online—limited non-verbal communication and social interactions.

Students and tutors reported a sense of disconnection when online, as they experienced less non-verbal communication and social interactions. Non-verbal communication (different types described in Table 4) was limited online because computer screens had to display the tiny windows of all students (usually 12) and the tutor. These windows tended to show the attendees’ faces only. Non-verbal communication was further reduced when students switched off their cameras or when a digital document was shared, which took up a larger portion of the screen. The weaker presence of non-verbal communication in online tutorials led students and tutors to feel less connected with each other (see non-verbal communication functions and related behaviors and quotes in Table 2).

Table 4 Types of non-verbal communication in tutorials, described by students and tutors
Full size table

Social interactions, in this study, include all student–student and student-tutor exchanges that were not related to the academic topic intended for the tutorial discussion. In this study, social interactions occurred before, during, and after the tutorial and during breaks in-between the tutorial. Conversational topics included jokes, discussing recent events, discussing other course work, technical problem-solving, sharing personal information and interests, and planning social or study activities outside the classroom. Students and tutors suggested that social interactions can lead to social connections. Students reported that it was more fun and easier to share their thoughts with friends than with strangers. Students also shared that they felt more “included” in face-to-face tutorials, because knowledge exchange was not the only purpose of their presence – they were also there for fun and making connections. Overall, students and tutors expect that students are better able to learn with and from each other when there is a sense of connection.

When online, there were barely any social interactions. Students and tutors attended just in time to start the online tutorial, turned off their cameras and microphones during the break, and left immediately after the tutorial. Tutor-L experienced the face-to-face tutorials as active, both in terms of the content discussions and social interactions. However, when the group had to move online, it became a quieter group – “a lot of the liveliness was gone, the jokes were gone, easy bantering was gone”, tutor-L described. Online tutorials were generally described as more business-like, focused, and formal. When face-to-face, tutors did not need to facilitate social interactions, because social interactions tended to occur naturally. When social interactions did not occur online, students appreciated the tutors’ efforts in initiating social interactions. However, these efforts did not work well because online social interactions felt forced. Responses to online social interactions tended to be short, involving only two or three students because students were aware that the others were listening in or waiting for them. Nonetheless, both students and tutors still suggested that tutors should continue initiating social interactions during online tutorials in an attempt to create a sense of connection, especially when there would be many online tutorials scheduled.

  • Theme 3: Student well-being suffered when tutorials were 100% online during COVID-19.

During the COVID-19 lockdowns, students experienced worsened well-being, which was noticeable during the online tutorials, also by the tutors. Students appreciated questions from tutors concerning their well-being during these online sessions. However, their answers tended to be short and “socially desirable”, because it felt uncomfortable to elaborate on personal issues while online.

Whenever face-to-face tutorials were possible, there was a higher chance for students to build their social network and organize social activities. “We had walks in between (class) to clear our minds,” student-L explained. Carrying out social activities, which was more possible face-to-face, was suggested by students to contribute to better well-being. For example, when students were facing hardships, such as having low motivation or emotional problems, having friends helped them to cope during the pandemic.

Still, there were students and tutors who hypothesized that well-being can be improved when students have the freedom to attend online tutorials from remote locations, because this may allow them to be with family or save on commuting time. However, some students reported that the time saved from not commuting to campus was spent studying instead. When students had to attend face-to-face classes again after the lockdown, Student-O shared that: “(Face-to-face) you get to see to know people and that expands your possibilities in terms of both having friends and academic possibilities. (With less studying time,) my grades have decreased, about 5%, but who cares? I’m happier… I’m happier to go to university. I’m happier to be around people. I’ve traded that 5% more for a nicer time at university. And I think it’s better. Because in the end, grades aren’t everything. You also need to have a richer life”.

3.3 Preference on scheduling face-to-face and online tutorials

In this section, we report on the final preferred schedules of students and tutors – schedules made for the sake of deep learning, social interactions, and the role of the tutor. The preferences are detailed in Table 5.

Table 5 Suggestions for a mix of face-to-face and online tutorials during a 2-month course with 14 tutorial meetings
Full size table

Overall, 15 out of 17 students and seven out of 13 tutors preferred a 100% face-to-face schedule. These numbers include two students and two tutors, who requested the option to go online only when necessary, retaining face-to-face as the preferred choice. Students and tutors emphasized that the option to go online should be strictly monitored, for example, only going online when there are emergencies such as a national strike of public transportation, natural disasters, and so on. Two students and five tutors preferred a mix of majority face-to-face schedule with some online tutorials. One tutor did not have a preference and chose a 50–50 mix of face-to-face and online tutorials. Reasons for having some online tutorials include:

  • the possibility to train online skills (i.e., online collaboration among students and for tutors to better their online guidance),

  • the possibility to promote inclusivity in discussions (through turn-taking),

  • the use of online facilities, such as having logistically convenient subgroup discussions in break-out rooms, co-creation (including quick sharing of information, digital whiteboards, and mind mapping programs), digital quizzes, and chat functions, and

  • for practical reasons (e.g., ability to have back-to-back online meetings and attend from remote locations).

Nonetheless, because social interactions were deemed to deteriorate online, students’ and tutors’ preferred limiting the number of online tutorials.

One pattern that stood out from the schedules suggested in Table 5 was that the first and final few tutorials were suggested to be face-to-face by all students and almost all tutors (except from Tutor-M). Considering the importance of building connections and trust, the vast majority of students and tutors preferred meeting face-to-face frequently, especially for the first and final few meetings of the course. Some students even suggested a non-academic session purely for team building at the start of the course. Students and tutors who preferred some online meetings requested interspersing the online meetings with face-to-face meetings.

4 Discussion

In this study, face-to-face tutorial meetings were perceived to deepen content discussions and create a sense of connection through social interactions and non-verbal communication. A 100% face-to-face schedule was the most preferred schedule by both students and tutors for reasons of cognitive, social, and teaching presence, which were operationalized as deep learning, social interactions, and the role of the tutor, respectively. Students and tutors emphasized the importance of social interactions to build connections and trust, and to stimulate discussions and stay motivated. Some students and tutors indicated that they would like to have the option to switch to online when necessary for practical reasons.

In terms of cognitive presence, there was a general degradation of content discussions in online tutorials, according to both students and tutors. With discussions that lacked depth, students were less able to learn from each other and to self-assess their understanding of the topic. We suggest that the online setting may present new challenges for students to be cognitively present. The limited processing capacity (also known as cognitive load (Paas & Van Merriënboer, 1994) that students have may be used differently in face-to-face and online meetings. In another qualitative study, when learning collaboratively via web conferencing (e.g., Zoom or Microsoft Teams), various elements of the online interface were perceived by students and interpreted by the researchers as extraneous cognitive load, such as irrelevant stimuli, different formats of information, discontinuous flow of social exchanges, and so on, which may not be experienced face-to-face (Çakiroğlu & Aksoy, 2017). In this study, several distractors were reported as well during online meetings, such as disturbing noises of the surroundings from different microphones and technical connectivity issues, which might cause extraneous cognitive load. Paas and Van Merriënboer (1994) proposed that the adverse consequences of high extraneous cognitive load misdirect the attention of students from relevant learning elements. When deciding on how to schedule face-to-face and online meetings, we need to keep in mind that the students should not be overwhelmed with any added extraneous load, whether face-to-face or online, so that their processing capacity is retained for collaborative learning.

In terms of social presence, the limited non-verbal communication and social interactions in online tutorials reportedly led students and tutors to feel disconnected from each other. Online tutorials tended to stay on-task (i.e., focused on the academic topic) – a phenomenon that was described in previous studies as well (Jonassen & Kwon, 2001; Lantz, 2001). When online, there were fewer social interactions with other members of the group, such as personal conversations, jokes, and others. This may contribute to the fact that online education tends to leave students feeling disinterested, isolated, and alienated (Rasheed et al., 2020). Previous studies have also reported that social presence was experienced less in online courses (Zhan & Mei, 2013). Situations in which students meet face-to-face or online may evoke a different sense of belonging. This may affect the students’ ability to participate, interact, and even disagree with others, while maintaining a feeling of comfort and trust. Importantly, feeling a sense of connection to other students has a strong and positive correlation with (Kozan & Richardson, 2014), and even predicts (Gutiérrez-Santiuste et al., 2015) students’ sense of cognitive presence. Feeling connected with others was also shown to positively correlate with course retention and final grades in online higher education (Liu et al., 2009). The underlying reason may be that group cohesion positively predicts student motivation, which positively predicts group productivity (Dolmans et al., 1998). Others hypothesized that the social aspects of collaborative learning stimulate the cognitive aspects of collaborative learning (Beuchot & Bullen, 2005; Clouder et al., 2006). Thus, social connectedness seems to be beneficial for collaborative learning.

In online tutorials, there was an apparent need for tutors to provide more guidance, as the online tutorials often stagnated and lacked depth. The founders of the COI framework assert that “cognitive and social presence, and ultimately, the establishment of a critical community of inquiry, is dependent upon the presence of a teacher” (Garrison et al., 1999). While teaching presence appears to be important in an online environment, many face-to-face collaborative learning settings such as PBL require the tutor to be less present to facilitate discussions, which are student-led and student-centered, and only stimulating the discussions when necessary (Hmelo-Silver, 2004). In the current study, tutor-interventions were considered to be necessary when online, while the flow of discussions between students appeared to be more seamless face-to-face. Online tutorials challenged tutors to guide students in a different manner (e.g., needing to be more directive), while the online setting also posed contribution obstacles for tutors (e.g., having less non-verbal communication). This finding aligns with a previous qualitative study, which reported that tutors needed to be more authoritative, positive, and didactic; in other words, to have more presence (Donnelly, 2013).

Students and tutors in the current study appear to have experienced less cognitive and social presence during the emergency remote learning during COVID-19, with teaching presence becoming more necessary online. The experiences reported in the current study align with a recent systematic review on student engagement in online learning during the pandemic (Salas-Pilco et al., 2022), where students experienced technological issues, challenges to develop soft skills, less exposure to effective teaching, and difficulties with self-regulated learning (Biwer et al., 2021). Yet, with time, students can also adapt to the new modality of online learning, through the usage of online platforms to support learning and development of digital skills. Moving forward, online learning can be better delivered with better accessibility to quality technology (having stable and strong internet connection, having noise-canceling microphones, etc.) and further training in digital skills for students and teachers. Should online learning become a necessity once more, higher education institutions should provide more support for social engagement and emotional well-being.

In the end, the majority of students and tutors preferred face-to-face tutorials, which were experienced with deeper academic and social exchanges. The preference for having sufficient cognitive, social, and teaching presence may be generalized to other education activities, as a recent meta-analysis of education across all age groups found that cognitive, social, and teaching presence had a significant positive effect on actual learning, perceived learning, and student satisfaction of learning outcomes (Martin et al., 2022).

The strengths of this study include sampling from students and tutors who had ample experience in both the face-to-face and online settings, enabling them to make comparisons of both settings. Students and tutors were also from the same faculty, thus having similarities in terms of their experiences and educational goals. A limitation of this study is that the experiences of online tutorials were from the COVID-19 pandemic. These pandemic-experiences may be different from non-pandemic times. Students have reported feeling more worried about pandemic-related problems, rather than focusing on their education. Students and tutors also suffered from the lack of social activities during the lockdown periods, leading to poorer well-being and less social connections. Furthermore, the move to online education during the pandemic for social distancing reasons was seen as a temporary situation by many. Consequently, less effort was put into optimizing online tutorials. In addition, the move to online education also caused a higher teaching workload, which affected the performance of many tutors. Therefore, the students’ and tutors’ experiences and opinions were probably influenced by these suboptimal conditions. Nonetheless, this study provides an initial exploration into the themes that may influence the experience of and preference for a certain mix of face-to-face and online tutorials.

5 Conclusion

In the end, the vast majority of students and tutors preferred a schedule of 100%, or at least a majority of, face-to-face tutorial meetings as the default option. Face-to-face tutorial meetings are perceived to deepen content discussions and create a sense of connection through social interactions and non-verbal communication.