Facilitating blended learning in teacher learning groups: experiencing social configuration during face-to-face and online meetings

From social constructivist learning theory, professional development is increasingly shaped in primary teacher education using teacher learning groups (TLGs; Vrieling-Teunter et al., 2024). TLGs are social configurations (such as teams, learning communities, or networks) in which in-service teachers, teacher educators, and student teachers, sometimes supplemented by researchers and/or experts, collectively learn through social interactions (De Vocht et al., 2023). In TLGs, this means that from a shared responsibility for the learning process and outcomes, solutions to practice-based educational problems are analysed, critically discussed, designed, and further shaped in the participant’s daily practice (Van Schaik et al., 2019).

Collective learning can be described as the conscious and active construction of knowledge through social interaction (Robbers and Vermeulen, 2018). It concerns retaining, deepening, and creating new knowledge considering the developmental process (De Jong, 2019) and involves collective learning processes that simultaneously include individual professional development (Robbers and Vermeulen, 2018). Collective learning can occur through structural cooperation based on a shared ambition (Castelijns et al., 2013). However, sufficient diversity within the group in which learning occurs is essential for participants to be exposed to different perspectives (Castelijns et al., 2013). Existing knowledge can then be deepened, and new knowledge can be created (Castelijns et al., 2013; Robbers and Vermeulen, 2018). The prerequisites for collective learning are a sense of interconnectedness and equality (Robbers and Vermeulen, 2018). Also, when participants are engaged and feel responsible for the collective process and outcomes, they act accordingly (Castelijns et al., 2013; Robbers and Vermeulen, 2018).

When the prerequisites of collective learning are translated to TLGs, the learning process can be described in terms of social configuration, comprising three dimensions: Shared Identity and Equal Relationships (SIER), Practice Integration (PI), and Long-Term Orientation and Goals (LTOG) (Vrieling-Teunter et al., 2022). The SIER dimension is about how TLG members work interdependently in equal relationships with a shared purpose and responsibility for collective success (Admiraal et al., 2016; Vrieling-Teunter et al., 2016). It refers to participants' interconnectedness, equality, and social interactions. The interrelationships, the feeling of safety, the degree of reciprocity, and the fact that all participants feel ownership of the learning process and show an active commitment to collective success are important. The PI dimension comprises the relationship between the knowledge shared and created in the TLGs and members’ daily teaching activities (Van Meeuwen et al., 2020; Vrieling-Teunter et al., 2016). This involves understanding how activities are integrated into participants' daily practices. The LTOG dimension concerns TLGs’ activities that focus on short- and long-term goals and reflect the TLG members’ social learning attitude (Meirink et al., 2010; Vrieling-Teunter et al., 2016). It contains the dialogue about and description of short- and long-term goals. Aligning activities with group goals is essential here.

Research on physical TLGs has shown that understanding and awareness of the learning process can help facilitate collective learning in TLGs (Vrieling-Teunter et al., 2022). By describing and characterising the dimensions of the social configuration, insight can be gained into how the learning process takes place within TLGs. Also, the description of the social configuration can serve as a starting point for reflection to determine whether it fits the predefined goals or whether adjustments are desirable so that the learning process can be optimised (Vrieling-Teunter et al., 2022). In this way, shared values can be made visible, and knowledge can be integrated into participants' daily actions (Vrieling-Teunter et al., 2019). The heterogeneous composition allows learning processes and outcomes to emerge that would not have been feasible for individual participants (Vrieling-Teunter et al., 2022). However, the participants' diversity and differences in experience and expertise require attention and facilitation when working together in equality to achieve shared goals (Castelijns et al., 2013). This means that the support provided by a TLG facilitator during activities, such as assisting with meeting planning and reminding teachers of prior agreements, plays a crucial role in fostering collective learning (De Jong et al., 2021). Despite the facilitators' crucial role in professional learning, there is a noticeable lack of literature analysing their function in TLGs beyond coordinating, organizing, or supporting participants regarding time, place, and intellectual capacity (Hadar and Brody, 2021). Studies rarely investigate facilitating activities in TLGs concerning learning and professional development (Patton and Parker, 2017).

Creating an appropriate learning and working environment that invites dialogues and encourages the development of collaborative skills is essential for collective learning in TLGs (Vrieling-Teunter et al., 2021). Traditionally, TLG learning and working environments are characterised by a physical space where face-to-face interaction and learning occur (García-Martínez et al., 2022). Yet, due to the increasing digitalisation and because of distance learning and working measures at the time of the COVID pandemic, TLGs are increasingly shaped in a blended learning format (García-Martínez et al., 2022; Last and Jongen, 2021). Last and Jongen (2021) define blended learning as the optimal combination of separate face-to-face and online learning activities that logically follow each other. They emphasise optimal alignment of learning activities, activating work forms, and the pursuit of maximum interaction in which information and communication technology facilitates learning activities (Bruck et al., 2022; Last and Jongen, 2021). Depending on the needs and goals of the learning activity, the specific ratio between face-to-face and online may vary (Last and Jongen, 2021).

Combining face-to-face and online learning environments can provide several advantages (e.g. Azorín and Fullan, 2022; Last and Jongen, 2021), ultimately increasing learning opportunities (García-Martínez et al., 2022). However, the effectiveness of blended learning depends on several factors (e.g. Van Dorresteijn et al., 2020). For instance, attention to the quality of the learning content in combination with the learning environment is an important factor, as well as careful consideration of the accessibility, design, and functionalities within the online learning environment (Last and Jongen, 2021; Yunusa and Umar, 2021). In addition, participants' technological knowledge and skills are important factors (e.g. Boelens et al., 2017; Thomas and Thorpe, 2019). Likewise, explicit attention to a safe learning environment and mutual trust relationships is necessary due to the physical distance (e.g. Van Dorresteijn et al., 2020). A certain degree of face-to-face contact is vital in building trust and creating a sense of belonging (Hanraets et al., 2011).

Further, it is essential to realize that the online context can be challenging as it decreases opportunities for social interactions and complicates communication, which may hinder a sense of interconnectedness; after all, it is more difficult to pick up non-verbal cues and signals (Kreijns, 2020). In line with this, De Vocht et al. (2023) found that it is more difficult to explore goals in-depth online. So, there are several challenges to facilitating collective learning in blended TLGs. Still, De Vocht and colleagues' (2023) study which took place during the coronavirus pandemic showed that blended TLGs have potential. Participants considered aligning goals and practice assignments to the varied face-to-face and online educational contexts, organising informal discussions, encouraging safety, equality, and connectedness, and organising meetings for the whole group as success factors (De Vocht et al., 2023).

Deploying blended learning requires thoughtful implementation and facilitation and cannot occur merely by copying activities and expectations from the physical learning work environment to an online context (Van Dorresteijn et al., 2020). It should also be taken into account that the online context offers fewer opportunities for social interactions (De Vocht et al., 2023; Kreijns, 2020; Last and Jongen, 202). This may affect the social configuration and specifically the SIER dimension, which involves social interactions and a sense of safety and equality (Van Dorresteijn et al., 2020). In this study, we aim to provide insight into the social configuration of blended TLGs by capturing participants’ experiences of both face-to-face and online TLG meetings by investigating the research question: How do participants experience blended TLG’s social configuration during face-to-face and online meetings? Understanding this can lead to recommendations to improve facilitation of collective learning in blended TLGs.

The present study was conducted at a primary teacher education institution in the Netherlands. Ten TLGs were active within this institution in the academic year of 2022–2023. During that year, eight face-to-face and four online meetings were scheduled. Three of the ten TLGs were selected for this study based on the completeness of the available audio data. The selected TLGs were heterogeneous in composition, including third-year student teachers (i.e. students), in-service teachers (i.e. teachers), teacher educators (i.e. facilitators) and experts (e.g. in the area of giftedness) and each had its own specifications (see Table 1).

Three participants per TLG were recruited for interviews using convenience sampling (Golzar et al., 2022): a student, a facilitator, and an external (e.g. teacher, or expert). Nine participants in total were interviewed, including two males and seven females, adhering to the guideline of interviewing between six and 12 participants to identify 90% of the codes (Guest et al., 2006). All TLG facilitators (of the 10 TLGs) participated in the focus group. The focus groups were held with the TLG facilitators because they had insight into the characterisation of the TLGs and were, therefore, able to share experiences regarding their considerations for organising blended TLGs. Ideally, a focus group consists of six to 12 participants (Del Rio-Roberts, 2016); in our case, the two focus groups consisted of 13 and 11 participants, respectively.

The qualitative data were provided by audio recordings of each meeting (14 in total) of the participating TLGs, nine semi-structured interviews with various members of the TLGs, and two focus group meetings solely consisting of TLG facilitators.

The first focus group aimed to obtain a characterisation of the blended TLGs (see Table 1) and gather initial social configuration experiences during face-to-face and online meetings. The second focus group discussed follow-up questions about blended social configuration experiences during the year. Example questions were: “How would you characterize your TLG?” and “How would or did you support SC during online meetings?”. The focus group interviews lasted approximately 40 min and were recorded using voice recorders.

A retrospective biographical interview is a form of narrative research design that provides insight into the experiences and perceptions of individuals (Creswell and Guetterman, 2021). A semi-structured interview guideline was developed for the interviews based on the Dimensions of Social Learning Questionnaire (DSLI; Vrieling-Teunter et al., 2019) supplemented by questions about BL which were pre-tested with a teacher not involved in the present study. Example questions are: “According to you, what are the goals of the TLG?”, “How did the online context influence the group atmosphere?” and “Did everyone contribute to the development of the joint product”? During the interviews, the interviewee was asked to draw a timeline and reflect on actual events during the TLG meetings. The interviews lasted a maximum of 60 min and were conducted and recorded via Microsoft Teams.

Audio recordings are a form of observation that can help establish actual behaviour, reveal unconscious behaviour, and prevent socially desirable responses (Baarda et al., 2018). To assess the extent to which the SC dimensions were observable during the meetings, audio recordings were made with voice recorders distributed before each TLG meeting.

The data were gathered according to ethical standards; participants participated voluntarily, and the research study was ethically approved by cETO the ethical committee of the Open Universiteit under number U2019/03249/HVM. Data collection took place from January 2023 until June 2023. During the TLG meeting in February, their respective facilitators invited participants from the three participating TLGs to participate in the individual interviews. The researcher then approached interested participants. All audio recordings of the focus groups, interviews and TLG meetings were transcribed verbatim.

The data were analysed using the data analysis software ATLAS.ti 23.4. Both deductive and inductive coding were used. For deductive coding, an analysis framework was prepared (see  Appendix 1²) in advance based on the dimensions of the social configuration (Vrieling-Teunter et al., 2022), and codes for blended learning were added based on blended learning theory (e.g. Boelens et al., 2017; Kreijns, 2020; Last and Jongen, 2021). The following inductive codes were added: ownership (social configuration -SIER), disadvantages blended learning, advantages blended learning, planning/timing, and tips/recommendations (blended learning). Sub-codes were added to the codes for the social configuration to make sense of the data by meeting type and experience, i.e. online, face-to-face, positive experience, or negative experience.

To increase reliability, two coders separately coded 10% of the data (O'Connor and Joffe, 2020) to determine the inter-rater reliability in the form of Krippendorfs alpha, of which a score between 0.67 and one is regarded as acceptable to perfect agreement (Krippendorff, 2018). The inter-rater reliability was α = 0.84, indicating a good inter-rater agreement. One researcher then coded the remaining data.

The findings are reported per SC dimension and describe both face-to-face and online contexts (see  Appendix 2³ for a coding overview per participating TLG and focus groups in online and face-to-face context). For each subdimension, the most characteristic quote from one of the qualitative data sources (audio, interviews, focus groups) is included to support the quantitative findings.

Overall, regarding the social configuration dimension PI, participants only expressed positive experiences both in face-to-face (n = 52) and online contexts (n = 20). TLG3 reported more experiences (n = 37) than TLG1 (n = 24) and TLG2 (n = 7), which aligns with the activities undertaken in the respective TLGs (see Table 1).

TLG1 (n = 3) and TLG2 (n = 6) only adjusted the products during face-to-face meetings. In TLG3, products were developed and adjusted according to feedback during both face-to-face and online meetings: “[during the] online [meeting] yesterday, it also came back. We adjusted the goals as we hadn’t included them yet, then more feedback was given” (interview_external_TLG3).

The focus group mentioned a positive online experience: “ … making appointments can be done very well and sometimes even better online because it is much more effective and much faster” (focus group_may_2023). TLG1 and TLG3 made appointments physically (TLG1 n = 7; TLG3 n = 5) and online (TLG1 n = 2; TLG3 n = 1) about the use of the products. TLG2 (n = 1) agreed on the timeframe for testing the final product in the face-to-face context.

Only TLG1 (n = 3) reached the point in the project where the product could be integrated into practice: “I addressed a big math question with the class to just let them spar” (interview_external_TLG1).

In the focus groups, participants regarded the discussions about practice experiences as positive in both face-to-face (n = 1) and online (n = 2) contexts. From discussing research outcomes, TLG1 only had positive experiences (n = 9) in the face-to-face context. However, the facilitator remarked, “It is a coincidence that this meeting is now a face-to-face meeting because it [discussing study outcomes] could also have been discussed online” (interview_facilitator_TLG1). This is exactly what happened in TLG3; they discussed their practical experiences online because of the planning of the meetings (n = 3).

Concerning long-term orientation and goals (LTOG), participants had more positive (n = 151) than negative (n = 9) experiences in the face-to-face context. TLG3 had no negative face-to-face experiences. Similarly, in an online context, participants had more positive (n = 75) than negative (n = 31) experiences. The biggest differences between face-to-face and online context can be found in the “relationship between goals and activities”.

Positive experiences resulted from determining long-term goals and objectives during face-to-face (TLG1 n = 4; TLG2 n = 6; TLG3 n = 7) and online (focus groups n = 1; TLG2 n = 2; TLG3 n = 1) meetings. Ambiguity about the long-term goals in the face-to-face context caused some negative experiences (focus groups n = 1; TLG1 n = 1): “We sort of live from meeting to meeting [without clear goals]” (audio_TLG1_physical_18_January_2023).

Conversations about long-term goals provided positive experiences in both face-to-face (focus groups n = 3; TLG1 n = 12; TLG2 n = 17; TLG3 n = 7) and online (focus groups n = 1; TLG2 n = 1; TLG3 n = 1) context: “Then we talk about where we are, where we need to go and when we need to be there” (focus group_January_2023). In contrast, determining the goal in the face-to-face context was experienced negatively in TLG1 (n = 1).

Once the goal was clear, aligning the goals and activities was perceived positively in both face-to-face (focus groups n = 2; TLG1 n = 29; TLG2 n = 31; TLG3 n = 33) and online (focus groups n = 8; TLG1 n = 17; TLG2 n = 24; TLG3 n = 19) context. Online meetings helped to keep each other focused and to keep deadlines short. All participants were aware that face-to-face and online activities differed and that it was important to think beforehand about the purpose of the meeting and whether goals could be realised: “I just know that we have less time online and that we should not do the same things online than we do face-to-face “ (interview_facilitator_TLG2).

Regarding the face-to-face context, TLG3 only had positive experiences because of the alternation among meetings and in-between working moments. Overall, negative experiences (focus groups n = 2; TLG1 n = 2; TLG2 n = 2) in the face-to-face context stemmed from the time spent in meetings. In the online context, the scheduling of the meetings caused negative experiences (focus groups n = 19; TLG1 n = 2; TLG2 n = 7; TLG3 n = 3): “The organization [teacher education institute] forced us to stick to the schedule, whereas if we had been allowed to choose for ourselves we would have said: at this moment meeting online is more convenient and at that moment it more convenient to meet face-to-face” (interview_facilitator_TLG2). All TLGs had to cancel one or more online meetings to replace them with a face-to-face (subgroup) meeting at another point. The reasons for this were a preference for physical interaction when working together on the product, as is possible during face-to-face meetings, and scheduling issues during students’ internship week.

The social configuration dimension of shared identity and equal relationships (SIER) was perceived positively more often in the face-to-face context (n = 230) than in the online context (n = 79). Negative experiences were equal for face-to-face (n = 40) and online (n = 39) contexts. The main differences between face-to-face and online context were observable in the dimensions “group atmosphere,” “sense of equality,” and “feeling of belonging”.

When meeting and working face-to-face, participants experienced the group atmosphere as positive (TLG1 n = 16; TLG2 n = 10; TLG3 n = 29); this was due to the participation in conversations, working in subgroups, completing assignments together, and visible involvement in the group, which was expressed by body language and facial expressions. Also, in TLG3, dyads conducted book reviews and shared experiences for discussion at every meeting. Some negative experiences in the face-to-face context (TLG1 n = 1; TLG3 n = 5) were caused by students who regularly showed a wait-and-see attitude: “They [students] want to contribute, but don’t always know what they can contribute” (interview_facilitator_TLG1).

Positive experiences (TLG1 n = 6; TLG2 n = 9; TLG3 n = 13) in the online context stemmed from using the chat function, working with an online collaboration tool, structured turn-taking, and in TLG3 because of the book reviews. Also, participants of the smaller TLGs (TLG1, TLG2) as well as participants working in subgroups (TLG3) experienced involvement: “At such an online meeting, everyone also gets to speak …. being in an online session with eight people, that just works” (interview_facilitator_TLG2). In some cases, involvement was regarded negatively online (focus groups n = 1; TLG1 n = 5; TLG2 n = 1; TLG3 n = 10). This was caused by low participation and technical problems (TLG3): “I had no picture at one point …. technical issues are distracting” (interview_external_TLG3).

In the face-to-face context, equality was experienced both positive (focus groups n = 1; TLG1 n = 8; TLG2 n = 27; TLG3 n = 8) as negative (focus groups n = 1; TLG1 n = 13; TLG2 n = 6. Occasionally, the feeling of equivalence between students and other participants was experienced as physically negative): “Sometimes you get the feeling that you lack some experience and some knowledge what they [externals and facilitators] already have “ (interview_student_TLG1). Negative experiences in TLG3 (n = 5) stemmed from unequal input from the subgroups.

Equal input from all participants and working in subgroups provided positive experiences in the online context (TLG2 n = 8; TLG3 n = 2). Negative experiences in an online context (TLG1 n = 3; TLG3 n = 2) were caused by the fact that facilitators were often talking: “But online it feels like there is less [equality] because the teacher educator (facilitator) is very dominant. She then gives explanations and sort of forgets about the rest” (interview_student_TLG3).

Participants (focus groups n = 1; TLG1 n = 13; TLG2 n = 6 n = 2; TLG1 n = 13; TLG2 n = 14; TLG3 n = 20) had a positive feeling of belonging in face-to-face due to the informal moments during the meetings and the body language and facial expressions: “You can also read a lot more body language the moment you are together physically which gives all kinds of information that you cannot get when you meet online” (interview_external_TLG2). A lack of connection between students and other participants caused negative experiences in the face-to-face context (TLG1 n = 4; TLG3 n = 5).

In online contexts, the feeling of belonging was perceived positively less often (focus groups n = 1; TLG1 n = 8; TLG2 n = 6; TLG3 n = 6) than in face-to-face context. However, participants indicated a feeling of connectedness during both types of meetings. Online meetings were used as an additional contact moment for the entire TLG. Also, connectedness with one’s subgroup was experienced through applications such as WhatsApp. Negative experiences (TLG1 n = 6; TLG2 n = 5; TLG3 n = 1) were caused by a stronger sense of distance from each other online: “There is a bigger wall between everyone, which makes it less easy to engage in discussions with each other … I have to make more of an effort to explain myself exactly online than if it were face-to-face” (interview_external_TLG1).

Perceived ownership was experienced positively in face-to-face context (TLG1 n = 2; TLG2 n = 5; TLG3 n = 8), which originated from expressing one’s own opinion, making one’s own choices and showing initiative: “ … I don’t remember what triggered me to start talking about it [choosing a book] … I just did it. Online I would have been less likely to do that” (interview_external_TLG3).

The positive experiences (TLG1 n = 1; TLG2 n = 3; TLG3 n = 5) in online contexts ensued from engagement in conversations and participants (TLG2 and TLG3) who showed initiative in setting up the online environment: “What I did very stubbornly … [I] created a folder for inspiration in Teams …” (audio_TLG3_physical_18_ January_2023). However, participants (TLG1 n = 2; TLG3 n = 2) showed less initiative and asked fewer questions during online meetings.

All participants described the atmosphere in their TLG as safe in both face-to-face and online contexts. The positive experiences in the face-to-face context arose from showing vulnerability and asking questions (TLG-n = 1; TLG1 n = 11; TLG2 n = 9; TLG3 n = 10). In the online context, the positive experience from participants of TLG2 (n = 3) and TLG3 (n = 3) was related to an introductory task: “I think that [the introductory task] was very important and that it created a good foundation for everyone to know that they can express their opinions, and dare to express them in both face-to-face and online meetings” (interview_external_ TLG2).

In the face-to-face context, participants (focus groups n = 2; TLG1; n = 6; TLG2; n = 13; TLG3 n = 15) experienced the reciprocity of the relationships as positive due to the conversations from different perspectives during the meetings. Positive experiences (TLG2 n = 2; TLG3 n = 3) in the online context stemmed from sharing information and the online work sessions: “We both had the microphones and camera on. I was doing my part, the student was doing her part, and in between we were sharing information and asking each other questions” (interview_facilitator_TLG2).

The present study investigated participants' social configuration experiences, comprising three dimensions: Shared Identity and Equal Relationships, Practice Integration, and Long-Term Orientation and Goals (Vrieling-Teunter et al., 2022), in blended TLG’s during face-to-face and online meetings. Participants experience social configuration during face-to-face meetings positively when they: (a) develop and adapt products (Practice Integration; TLG1, TLG2), (b) agree on the use of materials in (internship) practice (Practice Integration; TLG2), (c) discuss practical experiences (Practice Integration; TLG1), (d) describe goals (Long-Term Orientation and Goals; TLG1), (e) are able to see body language and facial expressions, and through the informal moments (Shared Identity and Equal Relationships; focus groups, TLG1, TGL22, TGL3), (f) see actively engaged participants (Shared Identity and Equal Relationships; TLG1, TLG2, TLG3) and (g) experience a sense of equality, reciprocity and safety (Shared Identity and Equal Relationships; focus groups, TLG1).

Participants experienced social configuration in face-to-face context negative when there is: (a) a lack of clarity about goals (Long-Term Orientation and Goals; focus groups, TLG1), (b) a lot of time for consultation compared to the amount of working time (Long-Term Orientation and Goals; focus groups, TLG1, TLG2), (c) low connectedness between students and other participants (Shared Identity and Equal Relationships; TLG1, TLG3) and (d) inequality between students and other participants (Shared Identity and Equal Relationships; focus groups, TLG1, TLG2) and between subgroups (Shared Identity and Equal Relationships; TGL3).

Online positively perceived social configuration stems from (a) being able to reach appointments faster (Practice Integration; focus groups), (b) discussing practical experiences (Practice Integration; TLG3), (c) using online meetings as an additional contact moment (Shared Identity and Equal Relationships; focus groups, TLG1, TLG2, TLG3), (d) promoting active participation by giving targeted turns and using the chat function and online collaboration tools (Shared Identity and Equal Relationships; TLG1, TLG2, TLG3) and (e) using applications in subgroups (TLG3) and the online learning environment (focus groups) to provide support (Shared Identity and Equal Relationships).

Online social configuration is perceived negatively due to (a) insufficient alignment of activities with the learning environment due to fixed scheduling of meetings (Long-Term Orientation and Goals; focus groups, TLG1, TLG2, TLG3), (b) low active participation (Shared Identity and Equal Relationships; focus groups, TLG1, TLG2, TLG3), (c) unequal input from participants due to project leaders mostly doing the talking during online meetings (Shared Identity and Equal Relationships; TLG1, TLG3), (d) low connectedness due to perceived distance and non-verbal communication that is difficult to see and interpret via a screen (Shared Identity and Equal Relationships; focus groups, TLG1, TLG2, TLG3), (e) barriers in interactions due to functionalities of the online learning environment (Shared Identity and Equal Relationships; TLG1, TLG2, TLG3) and faltering technology (Shared Identity and Equal Relationships; TLG3) and (f) lack of connection with the whole group when working in subgroups (Shared Identity and Equal Relationships).

In general, based on practice-aligned goals, learning activities can be chosen that ensure optimal alignment of face-to-face and online learning activities (e.g. Bruck et al., 2022). The present study shows that when activities are not appropriate to the intended purpose, insufficient alignment with the learning environment is experienced. An explanation for this could be the fixed planning of the type of meeting. According to Bruck et al. (2022), the sequence should not be fixed but should depend on the objectives. Boelens et al. (2017) recommend that in new groups, the first meeting should be organised face-to-face, focusing on goals and mutual expectations and laying a foundation for the group’s social cohesion. The present study shows that getting acquainted during face-to-face meetings provides a good start for organising subsequent face-to-face or online meetings.

In line with this, technology constraints hinder participation in online meetings. Good technical facilities and sufficient technological skills motivate and engage participants (e.g. Graham and Halverson, 2023). For online meetings, network bandwidth must be adequate, participants must have well-functioning hardware (e.g. for using images) and must be sufficiently skilled (e.g. to apply capabilities within MS Teams; Mintii, 2023). Attention to participants' expectations and skills at the start by the facilitator can guide how the meetings will be organised (Vermeulen et al., 2023).

For the social configuration dimension Long-Term Orientation and Goals, setting up blended learning environments where collective learning occurs is complex. It requires well-designed face-to-face and online seamlessly connected activities (e.g. Van Dorresteijn et al., 2020). Similar to previous studies (e.g. De Vocht et al., 2023) in the present study, working from a collective ambition and shared goals appears to guide activities. By formulating shared goals and ambitions at the start of blended TLGs, participants feel more connected to the TLG (Vrieling-Teunter et al., 2021) and there is a higher level of engagement (Bruck et al., 2022).

A prerequisite for collective learning in (blended) TLGs is the connection of ambitions and goals to participants' daily (internship) practice (e.g. Vrieling-Teunter et al., 2019). In the present study, the social configuration dimension Practice Integration was reported to a limited extent because, for TLG1 and TLG2, the topic did not sufficiently connect with participants' daily practice. This is consistent with findings from previous research on physical TLGs in which it was found that connection to (internship) practice can ensure a higher degree of engagement and connectedness with the TLG and is therefore more likely to lead to Practice Integration (e.g. Vrieling-Teunter et al., 2019). Also, for Practice Integration and Long-Term Orientation and Goals facilitation, it is essential that the face-to-face meetings ensure sufficient time for product development and working on long-term goals. As for online meetings, working on short-term goals and informing each other of the progress is good content. Here, it is crucial that the facilitators critically discuss with the TLG participants how face-to-face and online activities can best be organised to raise awareness and increase the values of collective learning and the quality of outcomes in blended TLGs.

When using blended learning, specific attention to the sense of community during online meetings is important (social configuration dimension Shared Identity and Equal Relationships) (Pei et al., 2023). A sense of community contributes to meaningful interactions, commitment to group goals, knowledge sharing, and support from other participants (Pei et al., 2023). The importance of creating an online sense of community is in line with the Community of Inquiry model (CoI; Garrison et al., 2000) which focuses on the online collaborative creation of meaningful knowledge by focusing on social presence (the 'authenticity' of individuals in online interaction) (Kreijns, 2020), cognitive presence (the way participants assign meaning and construct knowledge through reflection and conversation) (Ten Berge et al., 2022), and teaching presence (designing, facilitating, and arranging cognitive and social processes) (Pei et al., 2023). For online meetings of blended TLGs, this indicates the necessity to make online meetings more informal and personal (e.g. through a “check-in” at the start, telling anecdotes and humour (Boelens et al., 2017). In addition, activating participants during online meetings (e.g. through collaborative tasks and facilitating online (peer) feedback) can contribute to social and cognitive presence and thus increase interconnectedness, equality, and engagement (e.g. Pei et al., 2023).

The role of the facilitator in online collaborative learning is crucial (e.g. Singh et al., 2021). Teachers are used to setting up face-to-face learning environments where active engagement and interactions come naturally. This does not apply to blended learning environments (e.g. Singh et al., 2021). Teachers must have sufficient knowledge and skills regarding blended learning, be aware of online teaching pedagogy and didactics (e.g. Graham and Halverson, 2023), and be aware of their own beliefs regarding blended learning (e.g. Bruggeman et al., 2022). Teachers' beliefs about the effects of blended may cause adherence to traditional strategies (Singh et al., 2021), as shown in the present study where joint activities were conducted during face-to-face meetings and mainly practical issues were discussed online. For facilitators of blended TLGs, professionalisation regarding online teaching pedagogy and didactics and reflecting on one’s beliefs regarding blended learning could help optimise the online learning environment (Singh et al., 2021).

While social interaction is at the core of TLGs, the facilitator must also pay attention to conversations becoming more in-depth by promoting critical thinking and direction (Hadar and Brody, 2021). For example, exploring different perspectives is a substantial and beneficial component of the interactions between the facilitators and the participants, which might promote higher-order thinking in both face-to-face and online contexts. Although these efforts are more challenging online due to the decrease in social interaction (Kreijns, 2020), encouraging TLG participants to approach their practice with a questioning mindset, fostering the exploration of differing perspectives, and demonstrating an open-minded and reflective approach to teaching practice are key aspects of this discourse (Hadar and Brody, 2021).

The present study has some limitations from which suggestions for follow-up research arise. The first limitation is the small number of online meetings. All TLGs replaced one or more online meetings with face-to-face meetings or cancelled online meetings. As a result, the number of online meetings was much smaller than intended. Participants may have different experiences regarding the social configuration when there are more frequent online meetings. Therefore, the number of online meetings should be monitored in follow-up research on blended TLGs.

A second limitation concerns the method of sampling. TLGs were selected by purposeful sampling with the selection criterion being “the completeness of the number of audio recordings of TLG meetings”. This may have resulted in an insufficiently representative sample. Some caution in interpreting the results is thus in order. In follow-up research, taking a more representative sample (e.g. variation in duration and size of TLGs) may improve reliability.

The gender imbalance of the interviewees sample is a third limitation, with two males and seven females included based on convenience sampling. Several studies show nuanced variations in how different genders engage with blended learning context. For instance, studies indicate that female learners generally demonstrate more favourable attitudes toward online learning environments, whereas male learners often report higher levels of confidence in navigating and utilizing online platforms (Dang et al., 2016; Greier et al., 2022). Given these documented differences, future research on blended TLGs would benefit from a more systematic investigation of gender-specific issues, ideally incorporating balanced gender representation in the sample design. The fourth and last limitation concerns the study’s nature. Despite the rich data, only qualitative data was collected as a case study at one educational institution. Therefore, the results are not generalisable to other TLGs. Follow-up research at multiple educational institutions is needed to see whether the results are generalisable. A mixed-method design is also recommended in follow-up research.

Despite its limitations, this study provides interesting findings that can support facilitators of blended TLGs:

1. Ensure professionalisation regarding online teaching pedagogy and didactics and reflect on one’s beliefs regarding Blended Learning.

2. Technology can support online social interaction.

3. Provide autonomy in the planning of the type of meetings (online and face-to-face), depending on the goals

4. Plan the first meeting in-person for social cohesion and group formation.

5. Set long-term goals and targets to increase involvement and commitment.

6. Ensure practical experience exchange in face-to-face and online settings.

7. Activate participants through collaborative tasks in face-to-face and online settings.

8. Emphasize social presence in online meetings through interactive activities.

9. Promote critical thinking to raise awareness and increase the values of collective learning and the quality of outcomes.

The deductive codes are based on the DSLQ (blinded for review) and on literature about Blended Learning (BL; *Boelens et al., 2017; Kreijns, 2020; Last and Jongen, 2021). Inductive codes are in italics.