Amy-Louise Byrne
Tracy Flenady2†- 1School of Nursing, Midwifery and Social Sciences, CQUniversity, Sydney, NSW, Australia
- 2School of Nursing, Midwifery and Social Sciences, CQUniversity, Rockhampton, QLD, Australia
- 3School of Nursing, Midwifery and Social Sciences, CQUniversity, Cairns, QLD, Australia
- 4School of Nursing, Midwifery and Social Sciences, CQUniversity, Bundaberg, QLD, Australia
- 5School of Graduate Research, CQUniversity, Rockhampton, QLD, Australia
Purpose: This discussion article explores the pedagogical and institutional implications of cohort supervision in Higher Degree Research (HDR), challenging the traditional dyadic model. It aims to illuminate how cohort supervision can foster collaborative learning, enhance candidate wellbeing, support equity and inclusion, and democratize doctoral education. By examining diverse models and their implementation across global contexts, the article identifies key tensions and opportunities in transitioning to cohort-based supervision. It also considers how such models align with contemporary educational theories and institutional priorities, offering a timely reflection on supervision practices in a post-COVID academic landscape increasingly focused on inclusivity, efficiency, and scholarly community.
Design: The article adopts a discussion-based approach. The analysis is guided by Wegner’s theory of Communities of Practice and is organized around key themes: supervisory roles and authority, peer learning dynamics, identity formation, and institutional readiness. Through comparative insights, the paper highlights how cohort models are designed, facilitated, and experienced, offering valuable insights and discussion points.
Findings: Cohort supervision models offer significant benefits, including enhanced candidate engagement, reduced isolation, and improved academic identity formation. They promote collaborative scholarship and distribute supervisory responsibilities, but also introduce tensions around role clarity, peer dynamics, and institutional capacity. The success of these models depends on thoughtful design, structured facilitation, and robust institutional support. Challenges include inconsistent participation, expert dominance, and lack of formal policy. When well-supported, cohort models can improve completion rates and foster resilience. However, without strategic alignment and adequate resources, they risk becoming unsustainable or ineffective, particularly in digitally mediated or resource-constrained environments.
Conclusion: This article contributes a critical and timely rethinking of HDR supervision by positioning cohort models as transformative pedagogical strategies rather than mere logistical solutions. It advances the discourse by framing cohort supervision within democratic and collaborative learning paradigms, challenging entrenched norms of academic authority and individualism. By foregrounding the pedagogical potential of cohort models, it offers a fresh lens through which to view doctoral education; one that prioritizes community, reflexivity, and shared scholarly growth, and offers recommendations for institutions and supervisors.
Introduction
Supervision in research higher degree studies (HDR) has historically followed a master-apprentice model, privileging a dyadic relationship between a candidate and a supervisor. This traditional approach has long been valued for its capacity to deliver personalized guidance and foster deep intellectual engagement (Lee, 2008). Over the past two decades, the academic sector has experienced rising HDR enrolment numbers, increasing academic workload pressures, changing candidate demographics, and growing expectations for efficiency, collaboration, and inclusivity in doctoral education (Manathunga, 2012; McCallin and Nayar, 2012). Equity considerations are central to this shift, particularly for international candidates, neurodiverse learners, and first-generation HDR students who may face systemic barriers in traditional supervision models (Kover and Abbeduto, 2024). These broader shifts have prompted institutions globally to reimagine supervision models, with a particular focus on candidate wellbeing, timely completions, and the sustainability of supervision structures. This reimagining of HDR supervision has given rise to cohort supervisory models where groups of candidates are supervised collectively, often through scheduled group meetings, peer-led discussions, or interdisciplinary seminars (Dautel, 2020).
Cohort supervision has emerged as a notable innovation in HDR supervision and represents a move away from traditional dyadic supervision models to a more inclusive collaborative design (Wrigley et al., 2021). Cohort supervision refers to a structured approach in doctoral education where a group of candidates is supervised collectively rather than individually. It typically involves regular group meetings, collaborative learning activities, and shared supervisory support, creating a community of practice that fosters dialogue, reflection, and mutual accountability (Jacobs and Frick, 2024). Cohort supervision models differ from collaborative or co-supervision models, which may, for example, include industry partnerships as supervisors for a candidate (Pyhalto et al., 2023). A key point of distinction for cohort models is the collective approach to supervision for multiple students simultaneously.
The exact structure of cohort models may differ from discipline specific (de Lange et al., 2011), candidate, or supervisor led (Bista and Cox, 2014), and can include formal seminars, peer-to-peer mentoring, research circles, interdisciplinary forums, and facilitated workshops. Cohort models have been found to enable supportive and reflective practice, facilitating a sense of community and comradery (Barnett and Muth, 2008). In addition to promoting collaborative learning, these models are of great interest to organizations and individuals, as they are thought to offer efficiencies in time and resources (Kroll, 2016; van Biljon et al., 2020; Blake et al., 2022) while simultaneously providing access to expanded expertise. Although cohort approaches are gaining renewed traction, particularly in applied, professional and interdisciplinary contexts, there remains ambiguity around how these models are defined, implemented, and experienced by both supervisors and candidates (Carter-Veale et al., 2016; Blake et al., 2022).
In the Australian context, national bodies such as the Australian Council of Graduate Research (2021) and the Higher Education Standards Framework (Tertiary Education and Quality Standards Agency, 2021) have emphasized the importance of robust supervisory frameworks that foster quality, consistency, and candidate wellbeing, all of which may be supported via a cohort supervision approach. As such, cohort supervision is experiencing a revival, not just nationally, but across the United Kingdom, South Africa, New Zealand, and the United States of America (Carter-Veale et al., 2016; Govender and Dhunpath, 2011; Hutchings, 2017).
Despite increasing uptake and reported enhancements in candidates’ learning and sense of belonging (O'Neil et al., 2016; Wrigley et al., 2021), the implementation of cohort supervision models has not been without contention. Debates persist around their pedagogical soundness, the implications for supervisory authority and roles, the tension between collaboration and autonomy, and the preparedness of institutions to support such models systemically (Samuel and Vithal, 2011). These tensions are further complicated by the heterogeneity of cohort supervision structures, which vary in intensity, frequency, and leadership arrangements, often reflecting institutional priorities rather than pedagogical intent (Bista and Cox, 2014). As such, it is timely for supervisors, centres and academic institutions to reflect upon current and future supervision models, particularly in the context of an ever-changing post COVID-19 academic landscape. To facilitate thought and inquiry in this area, this discussion article presents examples of cohort supervision models reported in the literature, with the purpose of highlighting insights into key tension points. It also considers how institutions might strategically implement such models to further facilitate their benefits from a pedagogical and structural perspective.
This article adopts a discussion approach, synthesizing existing literature to critically examine the pedagogical and institutional implications of cohort supervision in HDR. Rather than presenting new empirical data, the methodology involves a selective but purposeful review of peer-reviewed articles, policy documents, and conceptual frameworks. Sources were identified through academic databases, and while not systematic in nature, carefully selected key words were used to identify sources focused on models of collaborative supervision and their reported outcomes for the purpose of synthesis and discussion. These included terms “Research higher degree,” “graduate student,” doctorate, master, masters, candidate “doctoral supervision” OR “higher degree research” and “doctor of philosophy,” along with “cohort supervision,” “collaborative supervision,” “research supervision” and “collegial model.” Embase, Informit, ERIC and Educational Source Ultimate were searched. No evaluation strategy of literature was adopted as this was a discussion paper only. Instead, articles were synthesized and used to draw out key features for debate and reflection.
The discussion itself was organized around key points within the data, including peer learning dynamics, identity formation, and institutional readiness, enabling a comparative and interpretive synthesis of diverse perspectives. This approach aligns with the principles of a discussion article, emphasizing conceptual clarity and critical reflection to inform future practice and policy directions.
Theoretical basis
This discussion paper leverages Wegners’s theory of Communities of Practice (CoP) (Wenger, 1998; Wenger et al., 2002) which provides a robust conceptual foundation for understanding the pedagogical value of cohort supervision in doctoral education, while also understanding the current limitations. The theory emphasizes learning as a social process situated within a community where members engage in shared practices, develop a collective identity, and construct knowledge collaboratively. Cohort supervision operationalizes these principles by creating structured spaces for three core principles: (1) Mutual engagement; (2) Joint enterprise; and (3) Shared repertoires of scholarly practice (Wenger, 1998). Through regular group meetings, collaborative workshops, and dialogic exchanges, candidates move from peripheral participation toward full membership in an academic community (Wenger et al., 2002). This transition fosters identity formation, epistemological awareness, and a sense of belonging, outcomes consistently reported as critical for doctoral success. By framing cohort supervision within CoP theory, the discussion positions these models not merely as logistical solutions to workload pressures but as transformative pedagogical strategies that cultivate scholarly communities and democratize learning.
Beyond the dyad: cohort supervision models
Traditional supervision models primarily involve a single or team of two academic mentors guiding a candidate through the research process. In contrast, cohort models are collaborative, offering intentional spaces for like-minded people to come together, and interact and learn though their community (Wenger et al., 2002). Cohort models may introduce complexity, often involving multiple supervisors, peer mentors, and group facilitators, raising questions about authority, responsibility, and role boundaries. These dynamics align with CoP theory, which frames authority as negotiated within a community rather than held by a single expert.
Burnett (1999) introduced one of the earliest and most detailed accounts of a collaborative cohort model in an Australian doctoral program. Burnett reported that while candidates reported feeling less isolated and better supported, the coordinating supervisor’s role was ill-defined, leading to tensions between academic oversight and facilitation. Findings highlighted the importance of clearly delineated responsibilities, particularly in models involving multiple supervisors. Similarly, Colbran (2003) also highlighted unforeseen challenges related to authority through a study that trialled a ‘supervision cell’ facilitated through e-learning tools. While candidates appreciated the increased transparency and access to feedback from multiple perspectives, the model raised concerns among faculty regarding the erosion of traditional supervisory authority. Some supervisors were resistant to the collaborative ethos, preferring to retain control over intellectual direction. This tension revealed how ingrained supervisory norms can inhibit the successful adoption of shared models, especially in discipline cultures where individual authority is paramount.
Garcia-Perez and Ayres (2012) added a unique dimension to this discussion following their use of collaborative workshops. They found that candidates often entered doctoral study with a tactical mindset, focused on completing tasks, rather than understanding research as a strategic, iterative process. They argued this limited perspective is frequently reinforced by traditional, transactional supervision. In contrast, they found the cohort model helped demystify the research journey by encouraging collaborative reflection, though expert dominance in group discussions occasionally inhibited candidates’ participation. As such, it is important to create space for dialogic exchange, not just content delivery, within cohort supervision.
The ‘Dissertation House Model’ espoused by Carter-Veale et al. (2016) was a more contemporary cohort supervision model trialled in the United States of America. The model engaged multiple mentors across disciplines, offering intensive, time-bound writing retreats. The House model was praised for its ability to scaffold writing progress, yet the authors noted the absence of formal policy frameworks to define supervisory roles. This led to inconsistencies in how mentors were selected and engaged, pointing to the critical need for institutional policy to support sustainability and fairness in supervisory allocation. Similarly, in Carr’s (2021) documentation of a model used for distance doctoral candidates in New Zealand, they highlighted the need for institutional structures that clearly define sustained coordination over time. Carr found that as candidates moved through different stages of research at varying paces, supervisory coordination became more difficult, with some supervisors struggling to maintain consistency in guidance, particularly when some candidates disengaged from the group.
Findings from these models underscore the risk of supervisory ambiguity increasing when cohort models lack embedded continuity plans. Indeed, role clarity is an essential element of successful cohort supervision. Govender and Dhunpath (2011) emphasised the importance of this, describing cohort supervision as a way to redistribute authority and foster candidate autonomy. However, they also reported tensions arising from conflicting advice given by different supervisors and peer networks, particularly when institutional or disciplinary norms around hierarchy and deference were not addressed explicitly. These tensions underscore the influence of broader HDR culture on how supervision is experienced, and the necessity of supportive, reflexive environments that empower candidates while guiding them through disciplinary expectations (Hutchings, 2017).
Taken together, cohort supervision can distribute supervisory load and broaden intellectual input, but also requires new thinking around role definition, accountability, and institutional support. A lack of clarity in roles and responsibilities, competing supervisory voices, and expert dominance can create confusion and inhibit candidate engagement, particularly when institutional frameworks are ill-equipped to support collaborative models.
Peer learning or peer pressure?
Cohort supervision models have been found to shape candidates’ autonomy in both enabling and constraining ways. On one hand, peer-based structures offer candidates opportunities to share experiences, troubleshoot challenges, and form a sense of academic community (Barnett and Muth, 2008). Carr (2021) particularly highlighted this in their work with distance doctoral candidates, showing that regular collaborative workshops helped ease the transition from coursework to independent research. However, they also observed that candidates progressed at different rates, and as the group lost cohesion over time, peer reliance became a limitation rather than an asset for some. More autonomous candidates were less dependent on group structures and were better able to sustain progress independently. This suggests that cohort models may not serve all candidates equally without deliberate scaffolding. Choy et al. (2015) offered similar observations; they found that the cohort model encouraged critical reflection and peer learning, especially when psychological safety was actively nurtured by facilitators. However, they also reported that some candidates hesitated to fully participate due to a perceived lack of challenge or a fear of being judged, illustrating that peer learning alone is insufficient without a carefully cultivated learning environment. Cohort models can mitigate isolation for underrepresented groups by fostering belonging; however, without deliberate facilitation, they risk reproducing inequities. For example, international candidates may struggle with language barriers, and neurodiverse students may require tailored communication strategies (Gumbo and Gasa, 2023; Australian Council of Graduate Research, 2021).
Concerns about engagement levels in peer-led doctoral work have been raised in discussions surrounding cohort supervision. Colbran (2003), in their work on e-supervision cohorts, noted that stronger candidates occasionally dominated discussions, unintentionally marginalizing less confident participants. This created disparities in engagement and progression unless active steps were taken to ensure balanced contributions. Some models have attempted to address these challenges to peer learning by redistributing leadership with the supervisory group. For instance, Dysthe et al. (2006) introduced a three-pronged model in a HDR program in Norway which combined supervision groups, candidate-led colloquia, and individual supervision. Their findings showed that while group-based supervision fostered a strong sense of community and exposure to multiple perspectives, some candidates deferred to dominant peer voices, delaying their own critical decisions and weakening analytical rigor. Similar dynamics have been reported in Australia, with Wrigley et al. (2021) reporting that the presence of strong peer groups created tensions where candidates deferred to consensus or relied too heavily on peer validation rather than asserting their own scholarly voice. This groupthink effect underscores the need for structures that support critical independence within collaborative environments. This is a risk that requires consideration and mitigation.
Despite some concerns around critical independence, there is evidence that the cohort model can have a positive impact on candidates’ confidence, sense of belonging, identity as researchers and time management (Barnett and Muth, 2008; Dysthe et al., 2006; Samara, 2006; Wrigley et al., 2021). Measured improvements include improved self-esteem and academic development (Colbran, 2003; Hutchings, 2017). Cohort models have also been shown to facilitate motivation and resilience, filling gaps that traditional one-on-one models sometimes leave unaddressed (Fynn and Janse van Vuuren, 2017; Wrigley et al., 2021). As such, cohort supervision models may support equity, diversity and inclusion for those at risk of isolation, through what is known to be a particularly challenging time for vulnerable groups, such as international candidates (Bager-Charleston et al., 2024). Indeed, where cohort supervision may have the greatest impact, is in its ability to bring together candidates from diverse backgrounds and research areas, supporting equitable engagement and success.
Peer learning within cohort models reflects CoP’s principle of mutual engagement, where trust and shared accountability underpin collaborative knowledge-building (Wenger, 1998; Wenger et al., 2002), but, of course, require careful consideration and skillful negotiation. Ultimately, peer dynamics within cohort supervision can be a double-edged sword, providing community and support, but also introducing risks of conformity, disengagement, or over-reliance on group validation. The degree to which autonomy is fostered or constrained depends not just on the candidates themselves, but on how the cohort model is designed, facilitated, and integrated into the broader doctoral journey.
Identity formation and engagement in collective scholarship
One of the most consistently reported benefits of cohort supervision is the development of academic identity and a sense of belonging for HDR candidates. Candidates in cohort models report that the structure builds motivation, trust, and collegiality (Wisker et al., 2007). These collaborative environments support identify formation through development of confidence, academic voice, and fosters a commitment to completion (O'Neil et al., 2016; Wrigley et al., 2021). Identity formation occurs not only through interactions with peers and supervisors, but also through industry collaboration, fostering a sense of relevance and future academic purpose. This aligns closely with Wenger’s (1998) CoP, which created shared repertoires of scholarly practice.
Peer learning is widely credited with reshaping how candidates approached the research process. Carr (2021) reported that in distance-based programs, cohort meetings became a critical site of emotional support and validation, helping candidates see themselves as legitimate members of the academic community. Candidates who initially lack confidence report feeling more comfortable discussing ideas and receiving feedback in group settings, which in turn promotes identity development and academic resilience. Certainly, cultivation of academic voice is a significant milestone for HDR candidates (Colbran, 2003; Garcia-Perez and Ayres, 2012). From their early conception, it was noted that participants in collaborative seminars were more likely to complete their dissertations, attributing this to the mutual encouragement and collective accountability fostered within the group (Burnett, 1999).
However, Garcia-Perez and Ayres (2012) highlight that when candidates perceive a lack of academic challenge, the cohort model risks being seen as a ‘soft’ option, underscoring the need to balance support with scholarly rigor. This is indeed an important point—engagement is not automatic or guaranteed. Choy et al. (2015) observed that when reflective dialogue was under-facilitated or when group dynamics were not actively managed, candidates withdrew or disengaged. Hutchings (2017) similarly reported that participation in online group supervision was inconsistent unless the sessions were well-structured and supported by institutional infrastructure, and those facilitating the groups. These findings emphasize that identity formation and engagement require active cultivation through thoughtful facilitation, psychological safety, and visible academic value. The risk of disengagement is real, particularly for vulnerable and isolated population groups, or for those who have poor digital literacy. Such risk must be considered and carefully managed to foster strong engagement.
Well-designed cohort models have the potential to create a shared scholarly space where both candidates and supervisors benefit. For candidates, cohort models can increase confidence, motivation, and academic belonging. Candidates move from peripheral participation toward full membership in scholarly communities, a process central to Wenger’s CoP framework (Wenger et al., 2002). For supervisors, they can enhance the visibility of candidate progress, improve critical engagement, and reduce the sense of supervisory isolation. However, the benefits depend on thoughtful design, structured facilitation, and institutional embedding. Where these are absent, the same models may lead to confusion, diminished accountability, and lower attendance, highlighting the fine balance between structure and flexibility in fostering meaningful engagement.
Institutional will versus institutional capacity
Cohort supervision models require significant logistical and administrative support, including travel arrangements, face-to-face workshops and intensives, and cross-sector mentoring structures. Without sustained institutional investment, the deficiencies commonly attributed to such models may be exacerbated (Choy et al., 2015). Therefore, the successful implementation of cohort supervision models is closely tied to an institution’s readiness and resourcing capacity. Institutional alignment, including clear policy frameworks, workload recognition, and appropriate infrastructure, are a fundamental enabler of sustainability, particularly in resource-constrained settings (Choy et al., 2015; Hutchings, 2017). After all, cohort supervision models emerged in response to a lack of resources – an issue which has become increasingly challenging for academic institutions.
Institutions must be willing to front-load support to realize longer-term benefits. This need was highlighted by Hutchings (2017) who investigated a professional doctorate program that used both face-to-face and technology-mediated strategies for cohort engagement. They found that when technological tools were well supported, online group supervision promoted inclusion, reduced isolation, and sustained scholarly momentum. However, when technical issues arose or when digital participation waned, candidates disengaged. Similarly, Colbran (2003) trialled an e-learning based model; while the model increased flexibility and transparency, it also exposed gaps in digital literacy among both candidates and supervisors. Combined, these findings stress the importance of institutional investment in robust and reliable digital infrastructure and ongoing support.
The lack of institutional policy guiding a model’s implementation and the absence of structured training has been shown to lead to uneven uptake and difficulties in sustaining participation over time. Fynn and Janse van Vuuren (2017) reinforced this point in a South African distance-learning context, where candidates relied heavily on non-academic support due to limited access to formal institutional resources. The authors argued that emotional and infrastructural support were often under-provided in traditional supervision structures. They further argued that while these types of supports may be increased via cohort supervision, the approaches require greater investment in awareness-building and connectivity to succeed. Indeed, for cohort supervision CoPs to thrive, institutions must provide resources and policies that sustain mutual engagement and shared enterprise (Wenger, 1998; Wenger et al., 2002).
Early cohort models were found to increase workload for supervisors (Burnett, 1999) indicating organizational resourcing must centre on the workforce implications. Subsequent models were careful to preface the need for time and institutional support, particularly in regard to administrative duties and the organizing and running of the model (Choy et al., 2015). However, structural misalignments and a lack of formal recognition for cohort leadership roles remain cited as key limitations to implementation of cohort supervision (Colbran, 2003; Samuel and Vithal, 2011). Nonetheless, studies have identified institutional gains when cohort models are properly embedded. At the individual level, cohort engagement fosters resilience and sustained scholarship, particularly when institutions invested in supervisory training and cohort facilitation. Garcia-Perez and Ayres (2012), for example, demonstrated how structured modelling workshops helped candidates strategically conceptualize their research journeys, while also offering experts a chance to reflect on their own pedagogical practices. These forums, when supported institutionally, served as spaces for mutual learning and innovation. Institutions with established processes for cohort facilitation (e.g., doctoral training centers) are better positioned to harness these benefits (Govender and Dhunpath, 2011).
Additional institutional benefits are also reported. Cohort supervision models may indeed increase HDR completion rates, strengthen academic progression and retention (Burnett, 1999; Carter-Veale et al., 2016; Dysthe et al., 2006; Fynn and Janse van Vuuren, 2017; Garcia-Perez and Ayres, 2012; Hutchings, 2017), a factor which institutions cannot ignore. Institutional readiness is not simply about having funding or technology, but about enacting coherent policy, embedding supportive infrastructure, recognizing supervisory labour, and fostering a culture that values collaborative pedagogy. Where these enablers are in place, cohort supervision models can offer significant returns in terms of candidate experience, research quality, and completion outcomes. Importantly, institutions must embed equity principles into cohort design, including accessibility measures, cultural competence training, and flexible participation modes to support diverse HDR candidates. This involves recognizing and addressing systemic barriers faced by international students, neurodiverse learners, and first-generation HDR candidates (Blake et al., 2022). For example, institutions may provide language support for non-native speakers, adopt universal design principles to accommodate varied learning needs, and ensure psychological safety within group settings. Cultural competence training for supervisors and facilitators can help create inclusive spaces where diverse perspectives are valued (Boghdady, 2025). Additionally, flexible participation options, such as hybrid or asynchronous engagement, are critical for candidates balancing caregiving responsibilities, employment, or geographical constraints. Embedding these measures not only promotes equity but also strengthens the pedagogical integrity of cohort supervision by ensuring that collaborative learning environments are accessible and empowering for all participants (Mahande et al., 2025).
Looking to the future of cohort supervision
The implementation of cohort supervision models in HDR programs represents a significant shift in academic practice, requiring deliberate organizational strategies to ensure success. Institutions and supervisors must view cohort supervision not merely as a logistical solution, but as a transformative pedagogical model. Indeed, cohort models represent a distinct pedagogical approach that reimagines how research learning is structured, delivered, and experienced. At their core, these models shift the focus from the traditional dyadic supervision model to a more collaborative, dialogic, and socially constructed learning environment. This aligns with contemporary educational theories that emphasize learning as a communal, iterative, and reflective process (Shi and Blau, 2020).
Pedagogically, cohort models foster peer-to-peer learning (Malfoy, 2005), leveraging the known outcomes of collaborative learning pedagogies such as shared inquiry, critique, and knowledge construction (Zhou et al., 2019). They are underpinned by a “democratic philosophy of teaching and learning” (Samuel and Vithal, 2011, p. 82). As candidates navigate the complexities of research together, they learn not only from supervisors but also from each other’s diverse perspectives, methodologies, and disciplinary insights. This multi-voiced supervision environment, as described by Dysthe et al. (2006), enhances epistemological awareness and encourages reflexivity. Importantly, cohort models also promote identity formation and a sense of scholarly belonging. By embedding candidates in a community of practice, these models can help demystify the research process and reduce the isolation often associated with doctoral study. Candidates begin to see themselves not just as learners, but as emerging scholars contributing to a broader academic discourse. In this way, cohort supervision is not simply a means to an end, it is a well-considered pedagogical strategy that cultivates deeper learning, academic confidence, and collaborative scholarship, when implemented with purpose, vision and clarity.
Cohort models are a deliberate shift from the traditional methods of HDR supervision, and the associated practices, methods and dynamics associated with them. As discussed, the benefits of implementing such models are numerous but the outcomes are contingent on robust institutional support and strategic alignment, as well as engagement from supervisors and candidates alike. In other words, the organizational approach to cohort supervision is a key driver in success.
Research indicates that cohort supervision can be successfully implemented with positive results. The Scottish Graduate School for Social Science (SGSSS; Dadau et al., n.d.) provides one example of how supervision can be embedded within institutional policy and practice to build candidate candidates. Their framework emphasizes structured planning, with supervisors and candidates agreeing on shared expectations for participation, feedback, and progression. The guide outlines practical steps such as scheduling regular group meetings, integrating peer review activities, and using collaborative platforms to sustain engagement. This cohort supervision model is positioned as a complement to individual supervision rather than a replacement, ensuring candidates retain access to personalized guidance while benefiting from collective learning. By formalizing these processes within policy, SGSSS demonstrates how institutions can operationalize cohort supervision to enhance scholarly community, reduce isolation, and improve completion outcomes while maintaining academic rigor (Dadau et al., n.d.). While certainly positive, the need to balance academic rigor with pastoral support, along with the need to manage frustration, anxiety and disappointment was also highlighted. Indeed, cohort models do not offer a silver bullet for supervision shortcomings but may offer value in many forms.
To mitigate the known issues around roles, responsibilities, workload pressures and additional administrative burden, institutional support and leadership is needed, allowing both candidates and supervisors to be supported (Wrigley et al., 2021). A lack of formal policy will act as a barrier to consistency and scalability (Carter-Veale et al., 2016). As such, institutions must develop guidelines that articulate the purpose, structure, and evaluation metrics of cohort supervision, ensuring alignment with graduate research strategies and quality assurance frameworks.
Likewise, institutions need also commit to adequate resource allocation which are pivotal to supporting consistency and growth. Successful cohort models have leveraged infrastructure that supports collaborative learning, such as virtual learning environments, seminar spaces, and administrative coordination (Colbran, 2003; Hutchings, 2017; Wisker et al., 2007). In this modern age, digital literacy and technology training is essential for candidates and supervision teams. Without these, cohort models risk becoming unsustainable or ineffective, with waning levels of engagement leading to attrition (Hutchings, 2017). Future research should examine how cohort supervision can operationalize equity, through inclusive pedagogies, universal design principles, and targeted supports for marginalized groups (Blake et al., 2022).
The recommendations advanced in this discussion article, such as embedding equity principles, clarifying supervisory roles, and investing in institutional infrastructure, are directly aligned with Wenger’s CoP framework (Wenger et al., 2002). For a community of practice to thrive, it requires intentional design that supports participation, psychological safety, and access to shared resources. Policy templates and facilitator roles ensure that mutual engagement and joint enterprise are sustained, while evaluation metrics provide feedback loops to maintain the integrity of the community. Similarly, equity-focused measures, including cultural competence training and flexible participation modes, enable diverse candidates to engage meaningfully in the shared repertoire of scholarly practices. These structural and pedagogical enablers reflect Wenger’s (1998) assertion that learning communities are not spontaneous, but are cultivated through deliberate organizational strategies.
Cohort supervision models represent a transformative shift in higher degree education. Future work must preface thoughtful design, policy alignment, and supervisor development to further develop, implement and scale such models. Through this discussion, it is recommended that institutions focus on:
1. Policy and governance: the development of formal policy, processes and standards which support cohort supervision, connecting to the principals of equity and inclusion, along with timely completion of degrees. The CoP theory may underpin such organization and oversight.
2. Roles and responsibilities: cohort supervision should be coordinated, managed, dynamic and ensure psychological safety. Supervision teams must be trained in cohort supervision and provide culturally competent communication and pastoral support.
3. Infrastructure and resources: digital and physical spaces are required to support cohort supervision, through hybrid, asynchronous methods which support inclusion.
4. Evaluation and sustainability: cohort supervision models must be sustainable, and engage in evaluation to ensure they are relevant, meeting candidate, supervision and institutional needs, while being careful not to preface institutional needs over others. Again, the core principles of CoP may facilitate evaluation and ongoing sustainability of such models.
Conclusion
As institutions face increasing demands for efficiency, inclusivity, and academic excellence, cohort models offer a compelling alternative to traditional one-on-one supervision. When implemented with strategic intent, these models can enhance candidate engagement, reduce isolation, and foster a collaborative research culture. However, their success is contingent upon clear organizational structures, adequate resourcing, and a strong pedagogical foundation. This discussion paper suggests that cohort supervision can be more than a logistical solution, but a transformative educational approach. However, such models are not without their challenges. Institutions must invest in policy development, supervisor training, and infrastructure to support sustainable implementation. At the same time, supervisors must actively adjust their own approach and expectations, finding new spaces between the tradition and the new. Cohort supervision models hold substantial potential to enrich the doctoral experience for both candidates and supervisors, provided they are embedded within a supportive, well-resourced, and pedagogically sound institutional framework.
Data availability statement
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.
Author contributions
A-LB: Resources, Investigation, Writing – review & editing, Conceptualization, Funding acquisition, Writing – original draft, Formal analysis, Methodology, Data curation, Visualization. TF: Writing – original draft, Funding acquisition, Investigation, Resources, Writing – review & editing, Formal analysis, Methodology, Validation, Supervision, Data curation, Visualization, Conceptualization. IW: Formal analysis, Methodology, Writing – review & editing, Investigation, Validation, Writing – original draft, Data curation, Visualization, Funding acquisition, Resources, Conceptualization. NN: Investigation, Conceptualization, Funding acquisition, Resources, Writing – review & editing, Writing – original draft, Data curation, Visualization, Methodology, Formal analysis. BZ: Methodology, Writing – original draft, Conceptualization, Formal analysis, Resources, Visualization, Data curation, Writing – review & editing, Validation.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the CQUniversity Cohort Funding Grant (2024). The funder had no role in the conceptualisation, design, analysis, decision to publish, or preparation of the manuscript.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that Generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
Australian Council of Graduate Research. (2021) ACGR good practice guidelines for quality graduate research supervision. Available online at: https://www.acgr.edu.au/wp-content/uploads/2024/07/ACGR_GPG_Quality_Graduate_Research_Supervision.pdf
Bager-Charleston, S., McBeath, A., Symons, C., and Charura, D. (2024). The importance of diversity and inclusion within the research supervision relationship: the views of research students and supervisors. Couns. Psychother. Res. 25, 1–19. doi: 10.1002/capr.12776
Barnett, B. G., and Muth, R. (2008). Using action-research strategies and cohort structures to ensure research competence for practitioners-scholar leaders. J. Res. Leadersh. Educ. 3, 1–42. doi: 10.1177/194277510800300101
Bista, K., and Cox, D. W. (2014). Cohort-based doctoral programs: what we have learned over the last 18 years? Int. J. Doctoral Stud. 9, 1–20. doi: 10.28945/1941
Blake, S., Capper, G., and Jackson, A. (2022). Building belonging in higher education. Available online at: https://wonkhe.com/wp-content/wonkhe-uploads/2022/10/Building-Belonging-October-2022.pdf
Boghdady, M. E. (2025). Equality and diversity in research: building an inclusive future. BMC. Res. Notes 18, 1–3. doi: 10.1186/s13104-025-07096-4,
Burnett, P. C. (1999). The supervision of doctoral dissertations using a collaborative cohort model. Counsel. Educ. Superv. 39, 46–52. doi: 10.1002/j.1556-6978.1999.tb01789.x
Carr, S. (2021). Transitioning the team: supporting distance supervised doctor of business administration students through collaborative online workshops. J. Open Flexible Distance Learn. 25, 1–13. doi: 10.61468/jofdl.v25i1.451
Carter-Veale, W. Y., Tull, R. G., Rutledge, J. C., and Joseph, L. N. (2016). The dissertation house model: doctoral student experiences coping and writing in a shared knowledge community. CBE Life Sci. Educ. 15:ar34. doi: 10.1187/cbe.16-01-0081,
Choy, S., Delahaye, B. L., and Saggers, B. (2015). Developing learning cohorts for postgraduate research degrees. Aust. Educ. Res. 42, 19–34. doi: 10.1007/s13384-014-0147-y
Colbran, S. 2003. Collaborative supervision of legal doctoral theses through e-learning. Retrieved from ResearchGate. Available online at: https://www.researchgate.net/publication/265006942_COLLABORATIVE_SUPERVISION_OF_LEGAL_DOCTORAL_THESES_THROUGH_E-LEARNING_1
Dadau, A., Smith, D., and Elliot, D. n.d. Value co-creation in Social Science PhD Supervision. Available online at: https://www.sgsss.ac.uk/files/Value-co-creation-in-Social-Science-PhD-supervision_SGSSS.pdf
Dautel, J. B. (2020). Applying a collective academic supervision model to the undergraduate dissertation. Psychol. Teach. Rev. 26, 18–26. doi: 10.53841/bpsptr.2020.26.1.18
de Lange, N., Pillay, G., and Chikoko, V. (2011). Doctoral learning: a case for a cohort model of supervision and support. S. Afr. J. Educ. 31, 15–30. doi: 10.15700/saje.v31n1a413
Dysthe, O., Samara, A., and Westrheim, K. (2006). Multivoiced supervision of master's students: a case study of alternative supervision practices in higher education. Stud. High. Educ. 31, 299–318. doi: 10.1080/03075070600680562
Fynn, A., and Janse van Vuuren, H. (2017). Investigating the role of non-academic support systems of students completing a master’s degree in open, distance and e-learning. Perspect. Educ. 35, 186–199. doi: 10.18820/2519593X/pie.v35i1.14
Garcia-Perez, A., and Ayres, R. (2012). Modelling research: a collaborative approach to helping PhD students develop higher-level research skills. Eur. J. Eng. Educ. 37, 297–306. doi: 10.1080/03043797.2012.684672
Govender, D., and Dhunpath, R. (2011). Student experiences of the cohort model of doctoral supervision. South Afric. J. High. Educ. 25, 525–538. doi: 10.38140/pie.v29i3.1698
Gumbo, M. T., and Gasa, V. (2023). Postgraduate supervision support in open, distance and e-learning: supervisors’ and key stakeholders’ views. S. Afr. J. High. Educ. 37, 123–140. doi: 10.20853/37-4-5318
Hutchings, M. (2017). Improving doctoral support through group supervision: analysing face-to-face and technology-mediated strategies for nurturing and sustaining scholarship. Stud. High. Educ. 42, 533–550. doi: 10.1080/03075079.2015.1058352
Jacobs, L., and Frick, L. 2024 The praxis of cohort supervision in a comprehensive open distance e-learning university: a conceptual framework J. Praxis High. Educ. 6 1177–1196. Available online at: https://www.jstor.org/stable/48736550
Kover, S. T., and Abbeduto, L. (2024). Toward equity in research on intellectual and developmental disabilities. Am. J. Int. Dev. Disabil. 128, 350–370. doi: 10.1352/1944-7558-128.5.350,
Kroll, J. (2016). What is meant by the term group mentoring? Mentor. Tutor. Partnersh. Learn. 24, 44–58. doi: 10.1080/13611267.2016.1165488
Lee, A. (2008). How are doctoral students supervised? Concepts of doctoral research supervision. Stud. High. Educ. 33, 267–281. doi: 10.1080/03075070802049202
Mahande, R.D., Abdal, N.M., Setialaksana, W., and Mappeasse, M.Y. (2025). Students’ perceptions of and preferences for equity in hybrid flexible learning modalities. J. Educ. Online. Available online at: https://files.eric.ed.gov/fulltext/EJ1407713.pdf
Malfoy, J. (2005). Doctoral supervision, workplace research and changing pedagogic practices. High. Educ. Res. Dev. 24, 165–178. doi: 10.1080/07294360500062961
Manathunga, C. (2012). Supervisors watching supervisors: the deconstructive possibilities and tensions of team supervision. Aust. Univ. Rev. 54, 29–37. Available online at: https://files.eric.ed.gov/fulltext/EJ968516.pdf
McCallin, A., and Nayar, S. (2012). Postgraduate research supervision: a critical review of current practice. Teach. High. Educ. 17, 63–74. doi: 10.1080/13562517.2011.590979
O'Neil, S., Schurink, W., and Stanz, K. (2016). The benefits of using small supervisor-initiated groups to supervise master's research: part 2-general articles. S. Afr. J. High. Educ. 30, 210–230. doi: 10.20853/30-4-613,
Pyhalto, K., Tikkanen, L., and Anttila, H. (2023). The more the merrier? PhD supervisors’ perspectives in engaging in co-supervision. Innov. Educ. Teach. Int. 61, 1460–1471. doi: 10.1080/14703297.2023.2258853,
Samara, A. (2006). Group supervision in graduate education: a process of supervision skill development and text improvement. High. Educ. Res. Dev. 25, 115–129. doi: 10.1080/07294360600610362
Samuel, M., and Vithal, R. (2011). Emergent frameworks of research teaching and learning in a cohort-based doctoral programme. Perspect. Educ. 29, 76–78. doi: 10.38140/pie.v29i3.1697
Shi, T., and Blau, E. (2020). “Contemporary theories of learning and pedagogical approaches for all students to achieve success” in Optimizing higher education learning through activities and assessments. eds. I.-S. Yukiko and T. McVey (USA: IGI Global Scientific Publishing).
Tertiary Education and Quality Standards Agency. (2021). Higher education standards framework. Available online at: https://www.teqsa.gov.au/how-we-regulate/higher-education-standards-framework-2021
van Biljon, J., Pilkington, C., and van der Merwe, R. (2020). “Cohort supervision: towards a sustainable model for distance learning” in ICT education. eds. B. Tait, J. Kroeze, and S. Gruner, vol. 1136 (USA: Communications in Computer and Information Science), 147–162.
Wenger, E. (1998). Communities of practice: Learning, meaning, and identity. USA: Cambridge University Press.
Wenger, E., McDermott, R., and Snyder, W. (2002). Cultivating communities of practice. USA: Harvard Business School Press.
Wisker, G., Robinson, G., and Shacham, M. (2007). Postgraduate research success: communities of practice involving cohorts, guardian supervisors and online communities. Innov. Educ. Teach. Int. 44, 301–320. doi: 10.1080/14703290701486720
Wrigley, C., Wolifson, P., and aMatthews, J. (2021). Supervising cohorts of higher degree research students: design catalysts for industry and innovation. High. Educ. 81, 1177–1196. doi: 10.1007/s10734-020-00605-3
Keywords: cohort supervision, graduate research, HDR supervision, higher degree, research
Citation: Byrne A-L, Flenady T, Wise I, Nijkamp N and Zupan B (2026) Cohort supervision and pedagogy in higher degree research: rethinking the dyadic model. Front. Educ. 10:1697992. doi: 10.3389/feduc.2025.1697992
Edited by:
Reza Kafipour, Shiraz University of Medical Sciences, IranReviewed by:
J. Zhang, Florida Gulf Coast University, United StatesAmanda Thomas, University of South Wales, United Kingdom
Copyright © 2026 Byrne, Flenady, Wise, Nijkamp and Zupan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Amy-Louise Byrne, YS5ieXJuZUBjcXUuZWR1LmF1
†These authors have contributed equally to this work