1 Introduction

The motivation behind this work originates from the reported lack of graduates with critical job-appropriate skills that are suitable for today's creative and knowledge-based economies. These extend beyond subject-level knowledge to encompass skills like creativity, communication, collaboration, critical thinking, technology literacy, and vocational relevance among others (Becker et al., 2017; Chamorro-Premuzic, 2019; Leung & Bentley, 2017). However, the existence of a “widening gap between higher education outputs and industry needs and expectations” (Pelletier et al., 2022, p. 47) with regard to young graduates’ soft and digital skills, is apparent across various fields (Alexander et al., 2019; Gürdür Broo et al., 2022; Karimi & Pina, 2021; Pelletier et al., 2023).

1.1 Background

The preparation of graduates with such a skillset is perceived as one of the major educational challenges of the 21st century, and an area in which universities are evidently unable to deliver (Mulgan et al., 2016; Thornhill-Miller et al., 2023). Insufficient communication, lack of relevance, uniformity, and concurrency between the industry and academia in terms of the objectives, approaches, and evaluation criteria that apply in each are the main reasons behind this skills gap (Karimi & Pina, 2021; Leung & Bentley, 2017; Oguz & Oguz, 2019; Pelletier et al., 2022; Turbot, 2015).This issue is particularly prominent in the Digital Creative Industries (DCIs) which are inherently linked to creativity, collaboration, and technology literacy as the key driving skills for innovation (Leung & Bentley, 2017; Nuccio & Mogno, 2023). The DCIs subsume all Design disciplines which, amongst others, include the fields of engineering, architecture, product, industrial, information, and interaction design (L. Dym et al., 2005; Pontis & Van der Waarde, 2020; Proctor-Thomson, 2013; Zimring & Craig, 2001). All of these aim to produce original solutions for real-world problems, adhere to human-centred philosophy, follow creative and iterative processes, rely critically on technology, involve technical and visual activities, and are heavily social and collaborative (Gabriel et al., 2016; L. Dym et al., 2005; Nguyen et al., 2016; Strobel et al., 2013).

1.2 Literature directions

In response to the aforementioned skills gap, recent literature suggests a paradigm shift toward the challenge-driven university, that addresses authentic problems, cultivates soft skills, and produces innovative outcomes (Chamorro-Premuzic & Frankiewicz, 2019). This shift can apparently be achieved through technology-enriched, cross-disciplinary, and cross-organizational (university-industry) collaborations, such as venture labs, entrepreneurial incubators, shared virtual talent networks, and curriculum-based industry partnerships (Pelletier et al., 2023; Pontis & Van der Waarde, 2020; Singh Dubey et al., 2022). As such, the recent years have seen a slow rise of University-Industry Collaborations (UIC), which are nonetheless, still far from achieving "wide penetration in higher education” (Becker et al., 2017, p. 12).

1.3 Research framework

Guided by these directions, this research proposes a robust type of university-industry collaboration, drawing from the theory of Communities of Practice (CoPs) to support its objectives, design, and outcomes. Originating from Situated Learning and Cognitive Apprenticeship, CoPs is a social learning framework that sees a group of people with common interests and goals in a given field, connecting with each other to learn together, as the situated members of a common practice (Brown et al., 1989; Lave & Wenger, 1991, 1999a). This work suggests that cross-organizational (university-industry) CoPs that are integrated into the HE curriculum, can narrow the above-mentioned skills gap and boost graduates’ employability and competitive advantage in the industry, by allowing rich university-industry interactions to take place early on in HE (Shreeve et al., 2010).

1.4 Research significance

This work extends CoPs with a cross-organizational dimension, to involve both members from the academia and industry (Kezar et al., 2017; Wenger et al., 2009). This proposition is grounded in evidence highlighting the enhanced social learning capabilities of CoPs, and their role in fostering professional identities across all disciplines in HE (DeChambeau, 2017; Gibson, 2019; Harvey et al., 2015; Irving et al., 2020; Jackson, 2016; Park, 2015; Pontis & Van der Waarde, 2020; Power & Armstrong, 2017; Pyrko et al., 2017; Tight, 2015; Townley, 2020). While some preliminary conceptualizations of the cross-organizational model exist, a sound theoretical framework to guide them, as well as, empirical evidence from their application and validation in real educational settings is still missing (Iskanius & Pohjola, 2016; Jackson, 2016). Importantly, there are no CoP studies with a focus on the particular socio-epistemic and technology demands, goals, and practices of a specific discipline (i.e. Design studies), versus a one-fits-all approach to CoPs (Amin & Roberts, 2008; Smith et al., 2017).

1.5 Research design & outcomes

This empirical work focuses on a pre-existing organic CoP of third-year university Design students, which was extended to involve industry members, namely, clients, alumni mentors, and evaluators. This allowed students and professionals to collaborate on industry projects based on authentic and vocationally-relevant requirements and practices, and receive feedback and evaluation from experts (Albats, 2018; Bhatnagar & Badke-Schaub, 2017; Ivascu et al., 2016; Lombardi, 2007).

The effectiveness of the cross-organizational CoP model was assessed by comparing the epistemic, creative, and social outcomes of the CoP participating—versus non-participating—students. Following this, a more in-depth investigation of the technology adoption, the learning value, and students’ identity transformation helped classify both the effective and challenging elements of the cross-organizational CoP model. Consequently, this research synthesized an affordable and transferable set of guidelines, divided into three components: epistemic structure, social setup, and technology configuration, that can address the particular needs of CoPs in HE Design Studies.

While the entire research corpus consists of five individual studies, focusing on the design, enactment, and validation of the cross-organizational CoP, this paper focuses on delivering guidelines to assist CoP stewards in designing, managing, and evaluating similar learning environments toward related objectives. According to Wenger et al. (2009) the term ‘steward’ may reflect educators, researchers, instructional designers, or technologists who act as CoP administrators.

2 Theoretical background

2.1 Communities of Practice (CoPs)

CoPs is a social theory that sees learning as occurring informally, by living and interacting through participation in a community, steering rich meaning-making processes and diverse opportunities for learning. The activities in these processes are seen to assume ‘meta-meanings’ that constantly evolve, defined as the ‘negotiation of meaning’ (Wenger et al., 2009; Wenger, 2010a).

CoPs have been extensively investigated with regard to their three ‘constituents of coherence’, joint enterprise, mutual engagement, and shared repertoire (DeChambeau, 2017; Johnston, 2016; Pyrko et al., 2017; Vangrieken et al., 2017). Joint enterprise refers to the common goals that CoP members pursue, mutual engagement refers to the active involvement and collective negotiation of meaning in the practice, and shared repertoire refers to a common vocabulary of terms, routines, and methods that members develop to speed up their practice.

Legitimate Peripheral Participation (LPP) is a concept reflecting partial, rather than full forms of participation, as a legitimate mode of belonging in the CoP. Specifically, observation and imitation of more competent others promote the gradual involvement and evolution of novices toward fuller forms of knowledge and competence. These bounded or internal forms of participation reflect the local dimension of participation in CoPs. (Lave & Wenger, 1999b; Wenger, 1998).

The global dimension in CoP theory refers to the participation of members in various other communities that may have boundaries but are still inter-connected, as “no community exists in isolation” (De Moor, 2015, p. 1). The social history and reified items of a CoP’s practice derive from both local and global activities, through the boundary encounters of their members and the reified objects that are transferred across practices, what is known as brokering (Cobb et al., 2018; Hefetz & Ben-Zvi, 2020).

2.2 Modes of belonging and identity in CoPs

While participation in a CoP persists even in states of inactivity, engagement denotes active involvement in serving the joint enterprise (Wenger, 1998). Engagement involves a “kind of personal investment that makes for a vibrant community” (Wenger et al., 2002a, p. 36) and together with imagination and alignment, they comprise the three modes of belonging in CoPs. Imagination refers to projecting one’s self across time and space, extrapolating from the lived experiences in the community, while alignment refers to the coordinated efforts of members to align with the practice (Wenger, 2010c). All three are constitutive parts of identity, which according to Pratt and Back (2013) is not merely the sum of knowledge and skills, but an ongoing process of becoming – a trajectory.

2.3 Virtual CoPs

Wenger et al (2009) provided a technology framework for virtual CoPs (VCoP) involving four components: a) tools, software used for specific purposes, b) features—particular tool functionality and properties, c) platforms—collections of various tools, and d) configurations—the entire technological setup comprising all previous components (E. Wenger et al., 2009).

The VCoP framework proposes that stewards need to first, define the orientations (typical activities) of the community and then map these against field-specific epistemic activities, to define a CoP’s technology configuration requirements.

2.4 Socio-affective interactions in CoPs

Socio-affective factors have received considerable attention in CoP, Design, and HCI research in recent years (Heuer & Stein, 2019; Sanches et al., 2019). Socio-affective interactions are emotional externalizations (i.e. familiarity, trust) in social groups that critically impact learning. For instance, the fear of exposing personal weaknesses and being judged by others in a CoP – known as the ‘Virtual Panopticon’ (Rayner, 2012)—can cause feelings of vulnerability and lack of intra-personal (self) trust in learners.

Exclusive remote CSCL and CSCW communities may present serious trust issues (intra and inter-personal), as well as conflict, competition, and lack of accountability, due to insufficient face-to-face interaction (Nilsson, 2019). The role of technology is thus crucial, in creating effective socioemotional CoP workspaces which can facilitate natural interactions and simulate active co-present communication (Stephanidis et al., 2019).

2.5 Design-oriented CoPs

Due to their nature, the Design disciplines are integrally connected to social and situated learning (Brown et al., 1989). According to research (Shreeve et al., 2010, p. 128), an experiential approach to learning is critical to Art and Design education, as it enables the enactment of an "artist, designer or performer". A few examples from literature where CoPs facilitate such approaches in Design education indicate that they can provide enhanced opportunities for collaboration, knowledge sharing, assessment (Eriksson et al., 2021), skills development, and transferability of expertise to other domains, while forming professional identities and connections to the industry (Gibson, 2019; Jackson, 2016; Morton, 2012).

For instance, with regard to experiential learning, Gibson (2019, p. 32) explored the relationship between making, crafts, and learning in CoPs, affirming that "successful practical learning relies on community participation and the sharing of common values and goals", where individuals seek to belong and produce innovative artifacts in result. Innovation can evidently be accepted, only if is valued by the CoP it adheres to. Similarly, working with experts from the areas of Fine Art, Graphic Design, Design for Performance, and Fashion Product Design, Shreeve et al., (2010, p. 135) studied the characteristics, spaces, and fundamentals of effective teaching and learning. They found that CoPs can facilitate an understanding of "how to be a practitioner" through socially situated types of meaning-making in the practice, especially when these evolve through legitimate peripheral participation. Students initially participate peripherally, learning from various levels of expertise, and gradually develop their own creative insights and designer identities in the process.

In architectural studies, researchers argued that at the core of all Design disciplines lies the studio, which enables students to socially enact the professional activities of their CoP, focusing on critiquing and shared knowledge-building (Morton, 2012). Evidently, studio-based practice is seminal for mediating real-world relevance in a community of scholars, as a primary component of the industrial practice (Adams et al., 2016; Harvey et al., 2015). Likewise, strong communities can develop around exhibitions or workshops in education, generating better opportunities for Design assessment since this can "include more voices and build greater capacity for student learning" (Eriksson et al., 2021, p. 141).

Pontis and Van der Waarde (2020, p. 241) challenge existing Design pedagogical methods and advocate for using "the world as a classroom", encouraging student participation in the broader professional CoP, tackling real-world problems, receiving feedback from peers and experts, and observing real-life workplace interactions, to deepen "the understanding of how professional practice interacts with theory".

2.6 Knowledge gaps in CoP literature

Lea et al., (2005, p. 1) described CoPs as “one of the most articulated and developed concepts within broad social theories of learning”, with multiple applications in education. Respective empirical interventions recorded the benefits of CoPs in fostering self-development—mainly through legitimate peripheral participation (Stone et al., 2017; Woo, 2015), enhancing reflective skills (Rourke & Coleman, 2009), evolving the ability to network and co-create knowledge (Allee, 2000; Hildreth & Kimble, 2004), and steering innovation (Goodyear & Casey, 2015) and professional growth (Khalid & Strange, 2016).

However, to date, CoP research largely reflects intra-organizational incentives, focusing on either academic or industrial contexts, while cross-organizational CoP research is still limited (DeChambeau, 2017; Stone et al., 2017). We argue that at times when university-industry alliances are increasingly endorsed by academic institutions, the CoP-supported link between education and practice is a critical step for innovation that cannot be overlooked.

Based on the critical role of technology in cross-organizational settings, it is also crucial to report on effective technology configurations based on empirical evidence from their adoption by members in blended or virtual CoPs in HE, with both collocated and remote memberships respectively (Spagnoletti et al., 2015).

Lastly, the bibliography lacks concrete governance mechanisms for CoPs of a specific social structure (i.e. small-scale, closed, public), scope (i.e. academic, industry, cross-organizational), and purpose (i.e. learning, professional development), within particular disciplines (Amin & Roberts, 2008; Smith et al., 2017).

3 Methodology

3.1 Research questions

Cross-organizational CoPs in HE curricula require a holistic approach—a CoP ecology per se – to their design and evaluation. Drawing from underlying studies attending to the individual components of this ecology, this work integrates the sum of its findings to address two overarching questions:

  • RQ1: What constitutes an appropriate social, epistemic, and technology configuration design for a cross-organizational CoP in HE Design studies based on empirical findings?

  • RQ2: What are the practical governance mechanisms for a transferable cross-organizational CoP model in HE Design studies?

RQ1 is addressed in Section 3 which discusses the design of the cross-organizational CoP's main components, while Section 4 provides a summary of findings from the CoP's enactment and evaluation. While a full account of the outcomes of RQ1 is available in peer-reviewed publications (disclosed), in this paper, these outcomes merely serve to support the overarching research aim, that is, to extract a CoP governance model, in response to RQ2.

3.2 Participants

This research involved 38 third-year students in a Multimedia and Graphic Arts course in a university of technology (disclosed) who enrolled in two consecutive 3rd-year Web Design and Development (WDD) modules (13 × 180-min lessons per semester). The students were divided by registration into the Multimedia and Graphics directions, forming the experimental (G1, N = 21) and control (G2, N = 17) groups respectively. These groups were selected by the instructor of the course who was also the primary researcher in this work since they shared identical program structures, syllabi, and GPAs over years one and two of their studies. As such, convenience sampling (Patton, 2002) was employed. Students self-formed teams of approximately four-to-five people in both groups.

Ethical permission to run the studies was obtained from the university’s department. The appropriateness of study methods was approved by an internal ethics committee. Further, all ethical considerations regarding data collection were addressed in signed consent forms by participants.

3.3 Research design

The entire body of this research followed a mixed methods (MM) multiphase design. Related evidence indicates that MM is highly appropriate for CoP studies,. Specifically, we sought to examine the effects of the cross-organizational CoP by measuring and comparing learning and creative outcomes through quantitative data and understanding the experiential dimensions of learning, through qualitative investigations of the emergent phenomena (B. Wenger-Trayner et al., 2019).

In this work, while both groups of participants followed precisely the same epistemic activities of the WDD module, only Group G1 actively participated in the CoP (experimental condition) (Section 3.2). Following a quasi-experimental research design for studies 2 and 3 (Table 1), we compared the groups' epistemic, creative, and social learning outcomes, and conducted more comprehensive qualitative investigations on the technology adoption, the value creation of learning, and identity transformation phenomena, working with the experimental group only, in studies 3 and 4. Study 2a was conducted for the purposes of validating the quantitative instrument used in Study 2.

Table 1 Multiphase research design: quasi experimental and qualitative study designs
Full size table

Additionally, two frameworks were employed to guide the analysis of data. First, the ACAD framework which aims to generate knowledge on the design of complex learning networks to support learning designers with proactive information shaped by particular technological (SET), instructional (epistemic) and social parameters (Fig. 1).

Fig. 1
figure 1

Horizontal (ACAD) and vertical (VC) approaches to analysis

Full size image

Aside of design, the ACAD framework (Fig. 2) was also employed as a horizontal approach to the analysis of the empirical data collected in this work (Goodyear & Carvalho, 2016). The second framework—Value Creation (Fig. 3) – was employed as a vertical approach to the investigation, aiming to gain a deep understanding of the value of learning in CoPs, attending to five dimensions (cycles) of learning (Wenger et al., 2011; B; Wenger-Trayner et al., 2019).

Fig. 2
figure 2

Epistemically, physically/digitally, and socially situated activities and outcomes

Full size image
Fig. 3
figure 3

Value Creation framework cycles

Full size image

4 The cross-organizational CoP ecology

The proposed CoP model draws from the Activity centered analysis and design (ACAD) methodological framework to define and analyze the components of its learning ecology: the social, epistemic, and technological (SET) components.

4.1 Social component

An organic, self-formed social CoP became evident in G1 early on in the research. This had emerged amongst students, driven by their common status, purpose, and interests, since year 1 of their studies. The CoP in G1 was then extended by the instructor (acting as CoP steward), to also include members from the industry. The CoP members were (Fig. 4):

  1. a)

    Instructor of the module

  2. b)

    Students

  3. c)

    Floating facilitator: final-year student acting as teaching assistant

  4. d)

    Alumni Mentors: alumni designers (three in semester 1, two in semester 2), with two-to-three years of professional experience who provided regular feedback on student deliverables.

  5. e)

    Industrial Experts: field professionals (three in semester 1, five in semester 2), with 6 + years of experience, responsible for evaluating the final websites, but were made accessible to students via a social network (SN) group from the semester start

  6. f)

    Industrial mentors (clients): five local-industry organizations who assigned student projects and provided related resources and feedback

Fig. 4
figure 4

Cross-organizational CoP social structure and levels of participation adopted from Wenger-Trayner’s (2011) participation model

Full size image

A full account of the various CoP roles and incentives is provided in published work (Mavri et al., 2020c).

4.2 Epistemic component

Five industrial mentors (clients) assigned different projects to different student teams in WDD-1, essentially, to design and develop websites addressing specific business needs (Table 2). Each project was developed twice, once by a team in the experimental (G1), and once by a team in the control group (G2). Both groups followed identical lesson structures, materials, and problem-based learning (PBL) approaches in class.

Table 2 Experimental and control group teams structure, authentic projects and industrial CoP membership
Full size table

The thematic areas of instruction and classroom-based processes were designed to coincide with respective project deliverables, to be reviewed by the CoP’s alumni and industrial mentors. This applied to G1 students only, whereas students in G2 engaged in their normal academic activities. In semester 2, the G1 students advanced their projects further, while sustaining their CoP memberships.

A full account of the learning and instructional design can be found in parallel work (Mavri et al., 2021a).

4.3 Technology (SET) component

The SET component reflects the physical, digital, or blended learning setting that hosted the learning activities of the CoP (Carvalho & Goodyear, 2014). In this work, the emphasis was on ensuring that the technology configuration facilitated vital Design-oriented CoP interactions (communication/collaboration/visual/technical activities). The configuration strategy was formed based on internal and cross-organizational objectives, like software/tool availability for all members (i.e. free/low-cost/subscription-based), suitability for WDD epistemic orientations (Section 2.3), tool familiarity (i.e. tools used by the organic CoP), and ease-of-use (i.e. conventional functionality) for all CoP members.

Figure 5 presents the resulting technology configuration, categorized into the ‘team-based’, ‘community-wide’, and ‘single-user’ contexts.

Fig. 5
figure 5

Design-based CoP Technology configuration

Full size image

In this work, with the exception of Google Drive, Docs, and Sheets, all tools in the Team and Single-user context are classified as Creativity Support Tools (CSTs) since they support Design activities that generate innovative outcomes (Gabriel et al., 2016).

Extensive reports of the CoP’s orientations, technology configuration, and adoption results are presented in parallel work (Mavri et al., 2020a).

5 Data collection, analysis, findings (all studies)

The entire body of this work gathered quantitative data from the WSCMI instrument (Zeng et al., 2009) (Table 6), Cumming et al.'s (2016) feedback coding scheme (Table 11), students' final exams scores (Table 7), system logs, and communication frequencies (Table 10), as well as extensive qualitative evidence, from pre, during and post-intervention focus groups, one-to-one semi-structured interviews, field-notes, SN group posts and chat logs (Table 3). Quantitative data underwent statistical processing and independent samples t-tests were conducted to examine mean differences between the two groups (Tables 8 and 9). Qualitative data were transcribed and analyzed employing either inductive thematic or priori codebook-driven analysis approaches (Table 11) (Braun et al., 2019; Saldaña, 2015).

Table 3 Data collection methods
Full size table

Qualitative findings from all studies extracted overall strong participation levels from – primarily – learners—and external CoP members through the abundant learning and creative collaboration exchanges that emerged from positive technology-adoption findings, in both within-team and community-wide contexts. Students were evidently “using technology to learn together” (Wenger et al., 2009, p. 41). They also valued their interactions with CoP experts as possessing significant real-world value, particularly focusing on the systematic feedback which urged them to rethink their work strategies, processes, and creative outcomes, without compromising quality and deadlines.

Findings also revealed that CoP technology configurations should uniformly facilitate various media channels for communication (audio/video/chat/share-screen), different user roles and privileges, and importantly, interoperability between SNs and productivity tools through common access and functionality.

Results also denoted a substantial shift in learners’ perceptions of achievement and reframed imperatives for learning and professional development. The CoP had evidently encouraged them to adopt a performance-driven—rather than a strategic (grades)—approach to learning. This signifies transformed learner identities into a pre-professional status (Jackson, 2016), driven by increased awareness of their prospective transition into professional practice.

In terms of quantitative findings, the evaluation scores of learners’ creative outcomes (websites) and conceptual knowledge gains (exams) revealed statistically significant differences in favor of the G1 (experimental) group, with a medium to large effect size (Cohen, 1992) (Table 8). Likewise, the communication frequencies in learner-instructor interactions of group 1 doubled the amount of those in group 2, denoting greater engagement in learning (Table 10). Lastly, feedback findings (Table 11) presented negative feedback as a significant predictor of lower creativity scores, and neutral feedback as positively correlated to higher creativity scores, suggesting that a moderated (plain-versus-negative) approach to feedback can yield improved outcomes in cross-organizational CoP contexts (Fig. 6).

Fig. 6
figure 6

Coding references charts

Full size image

A full account of the data collection and analysis processes, as well as, respective findings is published in various parallel studies (Mavri et al., 2020b; Mavri et al., 2021a, b).

6 Results: Design implications

Building on the research findings, this section provides a model, that is, a collection of actionable design guidelines, grouped under the three ACAD components: Set (technology), Social, and Epistemic. The themes and guidelines are supported by empirical evidence from the studies to help the reader gain better contextual understanding. Their purpose is to locate the effective and challenging dimensions that warrant attention during the design, governance, and evaluation stages of a CoP’s life.

Table 4 presents the three sets of guidelines in the cross-organizational CoP model and the following sections discuss these extensively.

Table 4 Design implications for Cross-organizational Design-based CoPs
Full size table

6.1 Set design implications

Based on the technology configuration (Set) and adoption findings, a generic guideline for CoP stewards is to:

  • SE1. Integrate member-preferred social networks (SN), field-specific creativity support tools (CSTs), generic productivity, and online showcasing tools in the CoP technology configuration

As explained (Section 2.3), technology configuration decisions should be informed by a solid understanding of the community’s orientations and cross-organizational needs, to cater for different roles and activities in the CoP. Evidently, tool availability, ease of use, and familiarity are key factors for the configuration, as CoP members do not typically “live in” field-specific tools—CSTs—or community platforms, for example. Empirical evidence indicates that their day-to-day activities typically transpire on productivity tools, communication/collaboration apps, and importantly, SNs. These should thus inform respective configuration design decisions.

The next sections present targeted guidelines for Design-oriented CoPs that rely heavily on practical (technical) and visual design communication in CSCL and CSCW settings.

6.1.1 Technical and design-oriented communication: Practical and socio-emotional considerations

Practical considerations

As it stands, technical communication in design-oriented CoPs relies on tools that facilitate reviewing and debugging of programming code in shared environments.

The configuration should thus employ tools with live-editing and code debugging facilities, that are visually differentiated from natural language text (i.e. conversational data). This is typical of technical Q&A sites, like ‘Stack Overflow’, that facilitate code-formatted snippets that can be edited and executed within the platform (Mamykina et al., 2011; Yang, 2016). Stewards can proactively incorporate such functionality through relevant APIs into the CoP’s platform (i.e. SNs) to:

  • SE2. Integrate effective technical Q&A interface capabilities, like code snippet sharing, execution, and debugging within the social CoP platform

The ability to tag posts can also help build a searchable ‘code’ index. Additionally, voting systems based on code correctness, generating scores and badges for users on their SNs, can help elevate interest and also provide statistics for evaluation purposes. Hence, the next guideline:

  • SE3. Integrate automatic or manual gamification features in the social CoP platform to promote student interest and engagement in the practice

It is equally important that members are encouraged to follow ground rules for natural and technical writing, as the following guideline suggests:

  • SE4. Guide members to make use of appropriate language for effective natural and technical communication in the CoP platform

Class-based PBL practices can train students accordingly; following short seminars on technical communication, students can practice their technical writing on specific tasks and undergo peer reviews based on their writing’s conciseness and communicational aptitude.

Socio-emotional considerations

The CoP’s platform should foresee the occurrence of socio-emotional phenomena that may hinder learning, particularly, for ‘underpowered’ learners who often tend to make unhealthy comparisons between themselves and peers. Such phenomena can be minimized by on-demand modular visibility for certain interactions, such as one-to-one, one-to-team, team-to-team, and team-to-community, during the practice. Implication SE5 below incorporates two sub-implications for: a) activity-driven and b) permissions-driven modularity:

  • SE5. Support modular visibility to accommodate various ad-hoc interactions, both from the initiating and the target member’s perspectives

The CoP platform/tool should offer selective initiator visibility. ‘Initiator’ is the person who starts an activity (i.e. question, task, artifact creation), and can do so by using their name, team name, or anonymously (Table 4). This prevents the exposure of personal weaknesses, where preferred, while sustaining flow in the practice.

On the other hand, the platform/tool should also afford selective target visibility, that is, making activities visible only to specific members. For example, a student can post a question to a specific classmate, an alumni mentor, and the instructor. Thus, communication becomes adaptive and fine-tuned to the members’ needs. The visibility feature can also include options such as activity and role permissions (Table 5). The interface should therefore:

  • SE5.1. Provide on-demand activity-driven permissions

Table 5 Modular visibility scheme: initiator & target visibility, activity & role permissions matrix
Full size table

This allows initiators to grant targeted users (i.e. collaborators) edit, view, review, and/or collaborate permissions, for a specific workspace (i.e. canvas#3) or a particular artifact (i.e. ‘top menu’). These permissions can bind to a preset index of user roles, to build customized access (Table 5) as per the following guideline:

  • SE5.2. Provide on-demand role-specific permissions

This helps minimize intrusion (i.e. overwrite, delete) in CSTs and other tools.

Both SE5.1 and SE5.2 guidelines aim to provide a combined matrix of initiator and target visibilities, linked to activity permissions (edit, comment, etc.) for specific user roles to facilitate the members’ ad-hoc needs. This creates a fluid, aggregated, multi-visibility, multi-activity, and multi-role environment, eliminating the need for separate tools (Table 5).

For clarification purposes, we offer an example scenario whereby an author (initiator visibility) can label an artifact under their name or anonymously. For target visibility, they can choose to assign review permissions for all alumni mentors, edit, chat, and voice-call permissions for team members 1 and 2, and view permissions for class-wide access. Likewise, a team leader can grant a specific target member view permissions for a large workspace area (i.e. entire team's canvas), and edit permissions for a particular artifact in that workspace.

Visual design interactions

The following guidelines focus on visual design interactions (i.e. 2D/3D drawing, diagraming, prototyping) that are inherent in Design-oriented CoPs. They mainly concern collaboration in synchronous CSTs and address the topic of workspace awareness (awareness of others’ presence and activities) in shared virtual spaces (Gutwin et al., 1996).

Insufficient workspace awareness can be the cause of overwrites, deletions, or duplicated work in shared CSTs (Forghani et al., 2014). This often generates an excessive sense of ownership, individualistic behaviors, mistrust, and lack of accountability in the CoP practice. Interfaces should instead facilitate true immersion and full awareness of shared activities in the CoP’s virtual spaces (Gabriel et al., 2016). The CoP’s technology should:

  • SE6. Aim to enhance workspace awareness with regard to the peers’ identity, position, and activity, particularly in visual CST workspaces

This should be intuitively ‘detected’ (lightweight information gathering) without added physical/mental overhead (Gutwin et al., 1996). In practical terms, CoP platforms should natively present others’ activity (i.e. displaying labeled user cursors moving in real-time). Additionally, artifacts’ state of edit can be labeled (i.e. ‘in-progress’, ‘completed’), to allow/prevent changes by others. Synchronous multi-channel communication (i.e. chat, voice, video conferencing, screen-sharing, remote connection) can also provide supplementary coordination support, according to the following guideline:

  • SE7. Integrate multiple channels for multimodal communication in the CST’s visual workspace

Interoperability

CoP participation should allow its members to work with spatially and relationally proximal elements, be it visual/technical artifacts, Q&As, messages, SN posts, calendars, and resources. This poses a number of governance challenges in CoPs. Managing various tools/platforms, for handling such elements:

  1. a)

    requires time and effort, being equally overwhelming for CoP stewards and members

  2. b)

    increases redundancy (i.e. repeated data in different tools).

  3. c)

    diffuses participation across tools, leaving some tools unpopulated or underused, known as “practice intangibility” (Probst & Borzillo, 2008, p. 343).

CoP stewards should therefore aim to:

  • SE8. Enable interoperability between CSTs, SNs, generic productivity, and other tools in the CoP’s technology configuration

On a practical level, this synergy requires technical knowhow for employing APIs to inter-connect applications. However, the CoP’s practice has a lot to benefit from a consolidated platform offering universal access, field-specific and generic productivity functionality (i.e. CSTs.), shared resources, and common login and navigation, ensuring spatial and relational proximity.

6.2 Social design implications

6.2.1 Power relations: Trust, competition, and accountability

Power equates the ability to define and claim knowledge in the practice; in short, power and knowledge imply each other in CoPs (Farnsworth et al., 2016). As power entails predominance in the meaning-negotiation processes, it is often contingent on the degree of participation and accountability in CoPs. Power asymmetries often surface in CoPs, both in their internal and external interactions. Cop stewards should:

  • SO1. Aim for an even distribution of power through the balance of trust, competition, and accountability in the CoP

Primarily, the power of external members (i.e. alumni and industrial mentors) in defining competence in the practice can be compromised due to insufficient contextual information on the internal practice. CoP stewards should:

  • SO2. Empower external CoP members with compound and in-depth information on their purpose and role, as well as about the other members in the practice

Adequate information like comprehensive requirements, expected academic outcomes, feedback focus, and rich contextual information (team setup, member roles and responsibilities, detailed communication plan, etc.) should be provided to external mentors, early on in the practice. Learners could also be responsible for supplying this information, overseen by the CoP steward, possibly, by transforming this into an assessed deliverable.

Likewise, power may unintentionally accumulate in the hands of a few members with more interest and motivation in learning and community-building intentions. Related theory posits that while a strong core group is necessary for driving community flow, it can also be steered to encourage—not intimidate—others. Next, we review guidelines that aim to manage such power, trust, and accountability imbalances, as the key social phenomena extracted from our analysis.

Interpersonal (peer) trust

Lack of interpersonal trust involves doubting others’ intentions, expecting opportunistic behaviors, and unhealthy competition (Hsu et al., 2007). This type of ‘moral hazard’ reflects some members well-intendedly sharing their work and others taking the opportunity to ‘steal’ ideas instead. Our findings suggest that early CoP-wide peer reviews on project work can help alleviate mistrust. CoP stewards are encouraged to:

  • SO3. Schedule regular synchronized deliverable reviews for peer teams commencing early on in the project cycle

A full-transparency approach lessens the chance of ‘copying’ work after it is publicly scrutinized and promotes honesty instead. It helps eliminate sustained secrecy—often followed by surprising revelations (i.e. hiding work until it's fully completed), a strategy practiced by competitive students. It also ensures that learners practice evidence-driven constructive and unbiased feedback, that requires sound subject-level knowledge and metacognitive aptitude, which are critical attributes for pre-professional identity development.

To avoid competitive comparisons and antagonistic behavior between teams, different clients and projects can be assigned to each team:

  • SO4. Assign different industry projects and clients to different CoP teams, ensuring that they require equal subject-level knowledge and technical competence

The instructor should ensure that different briefs are attuned to same-level requirements, regardless of their thematic foci. This also infuses diverse ‘industry-academia’ information streams into the practice, perceived as ‘highly interesting’ by learners.

Intrapersonal trust

Intrapersonal (self) trust—linked to self-efficacy—is influenced by interpersonal trust, both associated with relations of power (Broom, 2015). Self-efficacy refers to personal beliefs about one’s aptitude to perform and generate. More academically competent members typically present higher self-efficacy levels, sound awareness of their public status, and active participation in the practice. In contrast, previous embarrassing or intimidating events may discourage future participation. To mitigate such issues, stewards should:

  • SO5. Aim for mixed-competence teams as the CoP’s working subgroups

Teams generally tend to operate as entities of a collective identity. However, self-formed teams often end up with a one-sided (high/low) ‘accumulation’ of competence, since ‘sameness’ is favored, which is reflected in the CoP participation (Tereshchenko et al., 2019). Instead, mixed-competence/attainment teams can benefit learners on the ‘lower end’ through their interactions with more competent teammates, enhancing motivation and gradually empowering them to become full CoP participants.

Another way to foster inter and intra-personal trust to boost participation, particularly in blended communities, is to encourage sufficient face-to-face (offline) collaboration (Booth & Kellogg, 2015). Specifically, stewards should:

  • SO6. Aim for community-wide face-to-face interaction early on and throughout the life of the community in order to boost online participation

Co-located interactions can also support explicit and implicit knowledge co-creation in the form of accidental information ‘spill-overs’ between teams (i.e. during class, tutorials, break times), which are more unlikely to occur online. Likewise, trust has a lot to gain from geographical proximity between internal and external CoP members. Close proximity provides comprehensive social cues and helps contextualize the requirements, feedback, and overall communication between students, mentors, and experts. These can trigger more spontaneous behaviors on behalf of learners and encourage higher engagement levels in the CoP practice.

Accountability

Accountability is a constitutive component of CoP practice, driven by the members’ sense of joint enterprise and mutual engagement. This aligns with the degree of competence in the CoP (increased competence equals higher accountability). It is also intertwined with trust and power in the community and suffers more in exclusive online—rather than blended—CoPs (Nilsson, 2019). The following guideline can help improve accountability, avoiding one-sided power accumulation:

  • SO7. Limit the size of the CoP to enhance member accountability

Individualistic tendencies and lack of accountability, cannot be easily ‘hidden’ in smaller social groups with sufficient face-to-face time, as these tend to generate healthier ‘pressures’ of participation (Nilsson, 2019). By contrast, accountability can easily get diffused in larger or exclusive online communities. The next guideline follows Wenger’s (2002b) rationale for dividing larger communities into subgroups based on location or subject, to help accountability and participation:

  • SO8. Highlight the intended responsibilities of each CoP role at the beginning and regularly throughout the life of the CoP

Based on this work’s findings, the role, goals, and responsibilities of CoP members often get subdued by the complexity and obligations of everyday work-life and their multi-memberships across a landscape of practices (Wenger, 2014). A sustained scheme to remind members of their benefits, contributions, and responsibilities in the practice can boost accountability (Borzillo, 2017).

6.3 Epistemic design implications

Epistemic guidelines involve the design of tasks that guide the learning activities and outcomes (Goodyear & Carvalho, 2016). The three prominent themes under this component concern time, feedback, and the roles and purpose of mentors in the CoP.

6.3.1 Time

Time is fundamental in CoP-based learning. Wenger perceives the transformation of members’ identities as learning that happens “in time and space and identity itself is a time/space concept” (Farnsworth et al., 2016, p. 11). While time can evolve knowledge and competence in the CoP practice, it can also hinder learning, if not effectively managed.

An initial epistemic time/task plan, should be collectively decided between CoP members to guide the real-life CoP learning activities that ensue (Goodyear & Carvalho, 2016). All members should fully understand the time/task plan and align it to their practice, work styles, preferences, limitations, deliverables, and schedules. CoP stewards should:

  • EP1. Invite community-wide participation in the design of the learning ecology prior to its enactment

This step is crucial as it primarily helps educators, experts, and students co-define their learning ecology, which consists of tasks, people, tools, and places, as the key components of the practice (Novakovich et al., 2017).

In line with the ACAD framework, this process can be visualized through representations (i.e. network diagrams) to enable better comprehension. Aside from being inherently linked to the Design practice, these serve as reified artifacts to guide members throughout the practice. Stewards should:

  • EP2. Introduce visual representations to simplify the epistemic ecology and clarify its practical implications early on in the life of the CoP

It is equally important that this ecology is pilot-tested prior to its enactment, to uncover possible issues and allow for early co-configurations (collective adjustments) to serve the practice. It is thus reasonable to:

  • EP3. Allow for sufficient time to pilot-test the epistemic ecology prior to the commencement of critical CoP-based learning practices

Time management is significant in achieving sound epistemic outcomes through scheduled interweaving of the curriculum and CoP practice to ensure that their activities and objectives coincide. For instance, within the WDD context, academic lessons on GUI design should timely precede relevant deliverables (GUI prototypes) to be reviewed by alumni mentors. Likewise, practicing time-management methods (i.e. Gantt-charts) should precede the delivery of the project-plan from students to their industrial mentors, as the following guideline suggests:

  • EP4. Plan the academic curriculum to coincide – thematically and temporally—with CoP-based activities

6.3.2 Feedback

A primary goal of the cross-organizational model is to provide students with the experience of authentic industry feedback (from alumni and industrial mentors); hence the following guideline:

  • EP5. Aim for regular feedback and evaluation of student work from expert industrial CoP members to enrich the academic feedback processes

This process may naturally result in a large volume of comments, which are often ambiguous or conflicting, yet, representative of future real-life scenarios. Nonetheless, this work's findings indicated that feedback should be 'curated' to ensure that its focus, volume, and tone align with the epistemic objectives of the module. CoP stewards should:

  • EP6. Proactively negotiate the focus, amount and tone of feedback with external CoP contributors

Instructors should outline feedback guidelines in advance, to define the thematic focus (visual/technical), volume (word range, single/multiple mentor reviews per team), and tone (neutral recommendations). To encourage engagement, feedback should be articulated so as to invite student responses and be visible to the entire CoP for others' public reference, as the next guideline suggests:

  • EP7. Articulate comments appropriately to encourage reciprocal feedback for CoP-wide access

6.3.3 The purpose of expert CoP members

The role of external CoP members is critical in cross-organizational CoPs. This work has inferred that they can: a) enhance identification with the CoP’s joint enterprise, b) enrich the evaluation processes with authenticity through their feedback, and c) enable brokering by importing boundary elements from different CoPs (Wenger, 1998). The next recommendation applies to all external roles:

  • EP8. Invite industry CoP members with various degrees of expertise to provide briefs, insights, evaluation, and feedback to students

This guideline should be practiced in collaboration with the module/course instructor to orient external members toward the focus of the CoP.

Alumni mentors

Wenger (2014) posited that CoPs aim to decrease the distance between masters and novices, unlike traditional apprenticeship theories. Perceiving masters, as the ‘big figures’ can create a wide competence gap and compromise learning. Instead, mentors who are ‘only slightly ahead’ are more accessible for assistance. Novices can gradually co-create competence in practice, by negotiating more proximal meanings. CoP stewards should aim to:

  • EP9. Recruit recent graduates for the role of alumni mentors in the CoP

The word recent has both temporal and relational dimensions. The recency between alumni mentors and learners reflects relational proximity—how individuals relate based on affinity and similarity (Moodysson & Jonsson, 2007). Both roles share similar backgrounds, epistemic foci, and instructional experiences. On the contrary, the longer the time since graduation, the wider the relational gap, and the less the alignment in the ways of knowing (Wenger, 2010b).

Industrial experts

Rather than distant symbols, industrial experts should be accessible learners, to promote identification and cultivate the process of imagination and alignment with the global practice (industry). Novices should have legitimate access to the experts' backgrounds, trajectories (university-to-practice), and real work-life experiences, as highly valued inside-information (Morton, 2012). Stewards should:

  • EP10. Aim to share expert trajectories and ‘inside’ information about the industrial practice

This can become a transformative process of the learners’ identification with expert trajectories, to pragmatically ground them in both the favorable (achievements and successes) and unfavorable (challenges and burdens) events involved in the process.

Industrial mentors (clients)

Real-life clients are essential for cross-organizational CoPs, not just for assigning authentic projects, but also providing diverse industry information. Their feedback may not abide by Design terminology, and it is often messy, negative, or conflicting with theory. Learning to manage, counteract, and factor this back into the work, is an important mitigation skill for students. Community stewards should seek to:

  • EP11. Always include real industry clients and authentic projects to guide the CoP-based activities

6.4 The cross-organizational CoP model

The model comprises a total of three levels of detail:

  • Level 1: Summary-level (current section)

  • Level 2: Actionable guidelines by component & theme (Section 6)

  • Level 3: Extended version: includes guidelines associated with respective findings, bibliographic evidence, proposed instructional interventions, and evaluation measures for similar CoP-based ecologies, published online (Mavri, 2021).

All three levels provide structured guidance to educators and researchers for integrating cross-organizational CoPs in HE Design and similar studies (Fig. 7).

Fig. 7
figure 7

The Cross-organizational CoP governance model. Own work

Full size image

7 Discussion

This section discusses findings in relation to the research questions of the study as follows.

  • RQ1: What constitutes an appropriate social, epistemic, and technology configuration design for cross-organizational CoPs in HE Design studies based on respective outcomes?

This study addresses the lack of evidence on the design of cross-organizational CoPs in HE, responding to the call for CoP research that is localized to specific disciplines such as the Design and adjacent fields, versus a ‘one-fits-all’ approach (Hsu et al., 2007).

Five individual studies synthesized the entire body of this research, capturing, triangulating, and explaining the designed and emergent learning phenomena. These reported on a) the Design and critical interlocking of the technological, epistemic, and social components that comprise an appropriate ecology for the complex practices of Design-oriented CoPs (Section 4), b) the significantly enhanced (actual and perceived) learning outcomes (creative collaboration results, conceptual knowledge gains, feedback value), c) the positive shift in learners’ perceptions of achievement and their reframed Design-oriented aspirations, and d) the transformation of learners’ identities into their pre-professional and professional statuses (Jackson, 2016), encouraging effective industry transitions and enhancing employability.

As such, this work contributes a first-time design and validation of the cross-organizational CoP model to the growing community of researchers and practitioners involved in university-industry collaborations. It focuses on curriculum-integrated CoPs that aim to promote collaboration, creativity, and real-world vocational relevance in HE Design studies (Albats, 2018; Ivascu et al., 2016).

  • RQ2: What are the practical governance mechanisms as part of a transferable cross-organizational model for CoP-based learning ecologies in ΗΕ Design studies?

The second research objective concerns governance mechanisms for CoPs in specific epistemic areas in HE which aim to cultivate creativity, collaboration, real-world vocational relevance, and pre-professional identity development. Section 6 presents an exhaustive analysis of the three guideline sets of the governance model. Next we critically discuss each of these based on their contributions to and interactions with foundational CoP theory and related research.

7.1 Set guidelines

The Set component themes concern socio-emotional factors in technology-supported CoPs, as well as considerations for Design-based practices, specifically technical and visual (practical) communication. Interoperability is also an important dimension of this component.

Practical considerations concern field-specific functionality and usability guidelines, to enable efficient technical communication in SNs and visual interaction in CSTs (SE1). While these issues were formerly addressed by the literature (Dillenbourg et al., 2009; Gutwin et al., 1996), this work contextualizes them to CoPs. It provides empirical CoP evidence and guidance on facilitating multi-channel communication (SE7), multiple roles, permissions, and visibility options (SE5), and increased workspace awareness (SE6) across tools used in a CoP.

The Set component also addresses and counteracts previous findings on the resistive type of agency by anyone (human) and anything (non-human, i.e. tools & platforms) in the community (Novakovich et al., 2017). 'Non-human' forms of resistance reflect usability barriers that can hinder CoP practice, addressed by respective guidelines (SE2). ‘Human’ forms of resistance may be triggered by fear of vulnerability and exposure (Brass & Mecoli, 2011; Waycott et al., 2017), criticism (Baek et al., 2008), unhealthy self-comparisons (Crossouard & Pryor, 2008), and competitive or dishonest intentions (Chang et al., 2008), lack of authorial identity (Dennen, 2016; Waycott et al., 2017), and language/communication barriers (Huang et al., 2016), all previously reported in CSCL/CSCW literature.

The value of these guidelines lies in their potential to bolster the three constitutive dimensions of coherence in a CoP – namely: a) mutual engagement, that is, to support the full complexity of ‘doing things together’, b) joint enterprise, that is, to use tools that allow for this to be safely – yet—flexibly negotiated, and c) shared repertoire, that is, to expedite the development and effective use of a shared vocabulary (language, resources, patterns) amongst members of a virtual practice (SE3-SE4) (Wenger, 1998).

The guidelines aim to achieve this through a flexible, modular, and interoperable technology (SE8), integrating generic productivity, field-specific tools, and SNs to create the 'digital habitats' that CoP members can ‘live in’ during their practice (Wenger et al., 2009). These can facilitate narrow or wide modes of interaction, for team-based, community-wide (local), and public (global) settings, based on preference. In this way, they equally empower the local (learners, faculty, alumni mentors) and global (industrial experts, clients) CoP members to build and sustain agency in their practice.

From a theoretical standpoint, the Set component’s contribution is manifold. First, it actively enables Wenger’s (1998) critical local/global duality and acknowledges its balance as vital for cross-organizational practice. Secondly, it safeguards a powerful polarity in CoP theory (Section 2.3)—synchronous/asynchronous participation—which caters for “togetherness and separation across time and space” (Wenger et al., 2009, p. 14), allowing fluent meaning-negotiation processes, in real-time or asynchronously. Thirdly, it fully aligns with recent advancements in HCI research that value the affective dimensions of people-technology and people-people interaction, as critical in CSCL and CSCW settings (Heuer & Stein, 2019; Sanches et al., 2019).

7.2 Social guidelines

The Social component guidelines involve the management of power originating from the level of knowledge and competence, identification with, and accountability toward the CoP. This has a significant impact on the members’ levels of trust and competition.

Power relations are bound to emerge in CoPs. Yet this work, unlike Fox’s (2000), does not perceive them as ‘conflicts’, but rather, as asymmetries that may ‘silence certain voices’ in the local-to-local-to-global relations. These warrant attention from CoP stewards. As such, this set acknowledges the criticality of providing global members with in-depth information about the local practice. On the one hand, this grants them enough power to drive the meaning-negotiation processes, and on the other—as Fuller et al. (2005) posit—it allows them to benefit from the fresh insights of the CoP's novice members (SO2, SO6, SO8).

All the same, and in line with prior research which acknowledges the probability of local-to-local power asymmetries (Cundill et al., 2015; Smith et al., 2020), the model proposes ways to constructively avoid one-sided participation from a core learner group. This tends to dominate the meaning-negotiation and knowledge-creation processes in the CoP (SO1). Similar to educational and organizational CoP findings, this work locates the complex and uneven distribution of power in the trust structures of a CoP (Stroupe, 2014). It has therefore synthesized a subset of guidelines (SO3-SO7) that target both inter and intra-personal trust, to reduce competition, and support the peripherality of members in evolving into fuller forms of participation.

The cross-organizational model follows and strengthens Wenger’s (2013) call to design for balance in CoPs. The aim of CoP stewards, in this case, is not to demolish power. Instead, the aim is to prepare for it, prevent it from 'blocking' voices that deserve to be heard, and moderate it to encourage members in making their own claims to competence, creating healthy opportunities for learning (Wenger, 1998).

7.3 Epistemic guidelines

The Epistemic component themes reflect the criticality of time in the a) co-evolution of knowledge, b) the transformation of learners’ identities, c) the importance of feedback as a boundary object that enriches the local practice, and d) the boundary relationships between learners and experts in the CoP, which instill the three modes of belonging: engagement, imagination, and alignment, in learners, as prospective graduates.

With regard to time, the model takes into account the distinct thematic and temporal differences between instruction and CoP-based learning – a part-academic, part-professional process and – agreeing with Morton (2012) – it advises a form of organized synchronicity (EP4). In doing so, it concedes with foundational CoP theory which advocates the early involvement of CoP members in the design and testing of a CoP's epistemic ecology (EP1-3) (Wenger et al., 2002a). Moreover, it draws from the ACAD framework to propose: a) the collective needs analysis and careful planning of the ‘chain of operations’ that are likely to develop in practice, before the practice, and b) the creation of visual representations (versus abstract conceptualizations) to clarify the components and epistemic activities of the cross-organizational CoP, prior to its practice.

Equally important, the feedback guidelines (EP5-EP7) in this component constitute the backbone of the cross-organizational model. They respond to Boud’s & Falchikov’s (2006, p. 400) call for reconceptualizing “the place of assessment in learning beyond the academy”, that is diversified by the perspectives of “parties external to the educational institution”. While this generates rich learning prospects, the feedback that ensues is often complex and requires more intense metacognitive action and meaning-negotiations from learners (Novakovich et al., 2017). While in agreement with Jackson (2016) in that these actions are critical for students’ evolution into reflective practitioners, respective guidelines recommend moderating the feedback to leverage its full learning potential.

The cross-organizational model also foresees and cultivates two types of boundary relations in the CoP, distant and narrow, based on their level of proximity. Industrial mentors (clients) for instance, are epistemically distant to other members in the CoP since they themselves are not designers. Yet, they share ‘boundary’ information (i.e. culture, goals, perspectives) and reified objects (i.e. documents, vocabularies, formulas) from their own practices. On the other hand, narrow proximities reflect epistemically closer members (i.e. Design practitioners/scholars), these being, the mentors and experts in the CoP. Still, these may also have various sub-levels of proximity. Alumni mentors, for instance, are more proximal to undergraduate students, since they too are recent graduates. They are also proximal from a generational (age) and relational (perspective) outlook, as opposed to industrial experts. This follows Wenger’s (1998) foundational conceptualization of a CoP's learning value, which sees members with proximal epistemic and generational characteristics making smaller leaps of effort to learn together. By contrast, industrial experts – though epistemically proximal—reside further away on the generational and relational axis, and are thus more distant in the learners' day-to-day CoP practice (Wenger et al., 2011).

As such, the EP8-11 implications corroborate, enact, and expand the CoP theory through the cross-organizational dimension, by introducing a blend of multi-generational and multi-relational proximities that generate various boundary encounters in the practice (Culver & Bertram, 2017). The model aims to help learners form a mental matrix of the different entangled practices, roles, and competences involved in their wider CoP ecology. In agreement with Wenger and Trayner (2016), it also highlights other possible trajectories, helping them gain a sense of their own purpose and orientation within a landscape of practices.

The sum of these guidelines align well with Jackson’s (2016) work which focuses on the role of complex CoPs in the development of learners’ pre-professional identity early on in university. Identity thus begins its journey in its purely academic form and evolves into its broader professional dimension, through the rich boundary experiences gained in a cross-organizational CoP practice (Farnsworth et al., 2016; Novakovich et al., 2017).

7.4 Model transferability

The proposed governance model in this work provides targeted recommendations for educational CoPs, which are attuned to the socio-epistemic and technological dynamics of Design-based communities with a cross-organizational scope, to support the entry and epistemic development of learners in the Design disciplines (Amin & Roberts, 2008; Smith et al., 2017).

However, each CoP intervention may be subject to its particular intra-disciplinary conditions and characteristics. While different fields like architecture, computer science, and HCI, may not follow identical methodologies or tools, they share foundational similarities, making the guidelines readily adaptable to adjacent Design settings (Bhatnagar & Badke-Schaub, 2017; Zimmerman & Forlizzi, 2014). At the same time, the fact that our findings and guidelines confirm and align with those of others in the literature (Borzillo, 2017; Novakovich et al., 2017; Waycott et al., 2017), even within seemingly different genres of CoPs, reinforces their transferability into other domains and contexts. Nonetheless, further research and adaptations are needed for cases of distant transfer to leverage the model's full potential within specific disciplinary learning settings.

Lastly, to fully empower prospective CoP stewards with full clarity on how to adopt, steer, and evaluate the cross-organizational model in their own contexts, this research delivers both specificity and ample room for customization, to accommodate the practices of different disciplines. An extended online model version (Section 6.4) (Mavri et al., 2021) provides rich examples and ideas of practical instructional interventions, methods for measuring student performance and model effectiveness, and available technology tools, with a broad range of applicability to various CoP practices.

7.5 Conclusion

This work presents research that is interdisciplinary, drawing on diverse theoretical and empirical evidence from fields such as learning and instruction, social psychology, educational technology, Human-Computer Interaction, 21st-century soft and digital skills that are integral to Design education, as well as innovation strategies through university-industry collaborations.

From a theoretical perspective, it advances the CoP conceptualization into a cross-organizational CoP model, following its design, empirical enactment, and evaluation of its impact on learning in HE. It also employs CoPs as a robust social learning framework, to guide the university-industry collaboration initiative it proposes. This is reportedly missing from the literature since most of the existing UIC studies adopt frameworks from organization-based research to frame their design and evaluation (Albats, 2018; Etzkowitz & Ranga, 2015; Scandura, 2016).

From a practical perspective, this work integrates the abovementioned conceptualizations, theories, literature directions, orgamizational spheres and disciplines into a "coordinated and coherent whole" (Choi & Pak, 2006, p. 351). In doing so, it designs a robust learning ecology, analyzes and explains the emergent CoP-based learning phenomena, and derives practical governance mechanisms for use in similar incentives. Consequently, it offers an affordable and transferable cross-organizational CoP model to assist instructors, technologists, researchers, and practitioners who wish to adopt it to enhance their learning environments in HE.

Limitations

The research presents limitations with regard to external validity. Due to the small number of participants and localized nature of its sample (sample of convenience), it is difficult to generalize findings to the population of interest, that is, students in Design courses. That said, there is greater confidence in its ecological validity (i.e. transferring to different settings within adjacent disciplines), since the studies occurred in natural (classroom), versus controlled environments, the stimuli under investigation (i.e. websites, epistemic outcomes) were naturally occurring and concrete—rather than abstract and arbitrary. Additionally, the participants’ behavioral responses were representative of the real world, since the score-based tools employed (i.e. scales, questionnaires) are used extensively in real-life situations (Gouvier & Musso, 2014).

Future work

There is much room for enhancing the research’s trustworthiness and impact with future studies. Indicatively, while the Design disciplines share a high level of resemblance, further research into the learning environments of specific Design sub-disciplines can validate and augment the cross-organizational CoP model with more targeted information. Exploring cross-organizational CoPs in multi-disciplinary and diverse international or cultural settings can also shed light on the intricate socio-technical and socio-emotional learning dynamics of such CoPs. Lastly, the pandemic situation and similar circumstances have underscored the need to explore exclusively online cross-organizational CoPs, where physical presence is not possible.