Integrating Design-Based Research and Agile Scrum for Inclusive Educational Technology Design: Best Practices and Challenges from an Accessible Augmented Reality Learning Authoring Tool Project

Abstract

Designing inclusive multimedia systems that meet diverse accessibility needs remains a complex challenge requiring iterative refinement and sustained collaboration with end-users. This paper presents an empirically grounded case study of an accessible Augmented Reality (AR) authoring tool project that integrates Design-Based Research (DBR) with Agile Scrum. By blending DBR’s iterative design experimentation with Scrum’s adaptive software development cycles, we designed and implemented a multimedia authoring tool for instructors with diverse accessibility needs. We document the best practices derived through iterative evaluation, such as early stakeholder involvement, the use of accessibility-focused user stories, and the alignment of research evaluation cycles with sprint reviews. The study also identifies key challenges, including balancing research and development goals, coordinating interdisciplinary roles, and maintaining an accessibility focus under compressed iteration cycles. We situate our contribution within prior DBR–Agile integrations, emphasizing its novelty in addressing inclusive system design rather than pedagogical interventions. The paper offers actionable insights and a validated methodological framework for human–computer interaction (HCI) researchers and practitioners seeking to design accessible multimedia systems by leveraging the complementary strengths of DBR and Agile Scrum.

1. Introduction

Designing multimedia technology, such as augmented reality (AR), for inclusive education, that is, creating digital learning environments that are accessible, usable, and pedagogically effective for learners and instructors with diverse abilities, presents a complex and urgent challenge [1,2]. Educational technologies, particularly immersive multimedia systems, must accommodate a wide range of user needs—including visual, auditory, motor, and cognitive—without compromising either pedagogical efficacy or user experience [3]. Traditional, linear development methodologies often prove inadequate for this task, as they typically lack the embedded mechanisms for continuous user feedback iterative testing, and evidence-based refinement necessary to address the fluid and nuanced nature of accessibility requirements. In contrast, Design-Based Research (DBR) and Agile Scrum are two prominent methodologies celebrated for their user-centered, iterative frameworks [4]. This paper proposes and empirically evaluates a novel hybrid framework that systematically integrates DBR and Scrum to support the inclusive design of educational multimedia systems. The framework is demonstrated through the Accessible AR Authoring Tool Project, a case study focused on developing a tool that enables instructors—including those with hearing or visual impairments—to author AR-based learning content.

Design-Based Research (DBR) is an educational research paradigm focused on addressing authentic problems through systematic, iterative cycles of design, implementation, and analysis [5,6]. Rooted in pragmatic inquiry, DBR generates both a functional intervention and a set of design principles or theoretical knowledge grounded in real-world contexts. Its cyclical nature—involving continuous analysis, design, and evaluation—relies on close collaboration with practitioners and a blend of qualitative and quantitative methods to ensure the solution is both effective and theoretically sound [7,8]. However, DBR is also resource-intensive and challenging. Its iterations can be time-consuming and unpredictable, sometimes exceeding available time or funding [9]. Researchers must also be cautious of potential biases resulting from their deep involvement in the design process, which can compromise objective analysis [10,11]. Furthermore, because DBR studies are often context-specific, ensuring that findings generalize to other settings is an ongoing concern [12]. These challenges underscore the need for adaptive, time-boxed processes that maintain DBR’s research rigor while introducing operational discipline—an objective that motivates its integration with agile development methodologies.

Complementing this, Scrum, a widely adopted agile project management framework [13], is merged with DBR. Scrum is grounded on an empirical process control model that emphasizes iterative development, frequent delivery of working software, and continuous stakeholder feedback [14]. Scrum organizes teams into time-boxed sprints, governed by specific roles (e.g., Product Owner, Scrum Master) and ceremonies (e.g., daily stand-ups, sprint reviews) to enable rapid adaptation to change [15,16]. For educational software, Scrum’s focus on demonstrable, working prototypes and responsiveness to emerging user needs can effectively align the engineering process with evolving pedagogical requirements. However, Scrum by itself does not prescribe mechanisms for theory generation or educational evaluation, limiting its capacity to contribute to scholarly knowledge production.

Despite their distinct origins—DBR from educational research and Scrum from software engineering—the two methodologies share a fundamental commitment to iterative design and continuous stakeholder involvement [4,17]. This common ground suggests a powerful synergy: DBR provides the rigorous research framework to ensure the educational intervention contributes new knowledge and is contextually valid, while Scrum offers the structured development framework to efficiently manage the engineering process and deliver functional prototypes within bounded cycles [17]. When integrated, these methodologies can combine DBR’s theoretical depth with Scrum’s operational efficiency to advance both research and practice.

However, integrating these frameworks raises critical methodological questions for practitioners and researchers: (1) How can the data collection and analysis activities of a DBR cycle be effectively integrated into Scrum’s sprint-based cadence? (2) What mechanisms are needed to sustain a strong focus on accessibility and inclusion during rapid iterations of agile development? (3) How should the roles and responsibilities of researchers, developers, and end-users be coordinated within this hybrid model?

This paper addresses these questions by documenting our experience in the Accessible AR Authoring Tool Project, where we explicitly integrated DBR and Scrum to design an inclusive learning authoring tool. The project’s central goal was to empower non-technical instructors, including those with disabilities, to create engaging AR learning experiences independently. Over multiple DBR cycles, we collaborated with instructors and students to identify accessibility needs, implement specific features, and evaluate the tool in authentic educational settings. The study’s novelty lies in the articulation and empirical validation of a hybrid DBR–Scrum framework, specifically tailored for inclusive multimedia system design, which advances inclusive educational technology by balancing research rigor, development efficiency, and accessibility-centered co-design. Throughout this process, we documented the best practices and challenges to derive methodological guidance for similar research contexts.

In the following sections, we first establish the methodological foundations of this work by providing an overview of DBR and Scrum. We then review the related literature on hybrid methodologies, highlighting the gap in inclusive multimedia design research that this study seeks to address. Subsequently, we detail our methodological integration of DBR and Scrum in the AR project, including our use of accessibility-focused user stories and adaptive sprint planning. Next, we present the best practices and challenges distilled from our experience, offering concrete strategies for mitigation. Finally, the last section concludes by summarizing the theoretical, methodological, and practical contributions of the study and outlining directions for future research in HCI and accessibility-centered learning design.

2. Background: Methodological Foundations

2.1. Design-Based Research (DBR) for Multimedia Systems

Design-Based Research (DBR) emerged in the early 2000s as a powerful methodology for bridging the gap between controlled laboratory studies and the complexities of real-world educational contexts [18]. Instead of isolating variables, DBR engages directly with authentic learning environments to generate both practical solutions and theoretical insights [6,19]. A classic DBR project involves multiple iterative cycles: researchers identify a problem in practice, design an intervention (such as an educational game or an AR authoring tool), implement it in collaboration with practitioners, and systematically collect and analyze data to refine the design [7,20]. Through these cycles, DBR generates design principles and localized theories about learning that are grounded in empirical evidence.

Key features of DBR that are particularly relevant to the development of multimedia systems include [5,21,22]:

• Collaborative and Design: Researchers and practitioners (e.g., instructors, students and designers) work together in all phases, ensuring that the intervention is theoretically grounded and practically viable in its real-world context.
• Iterative Refinement: The design of a multimedia artifact evolves across cycles; early prototypes are treated as exploratory tools for empirical learning rather than as final products, enabling the discovery of design principles through repeated testing.
• Mixed-Method Evaluation: Qualitative observations, interviews, and quantitative measures (e.g., usage analytics, usability data) are combined to evaluate the effectiveness and user experience of each iteration.
• Theory and Principle Building: DBR seeks to contribute both design principles and mid-range theories that guide the creation of technology-enhanced learning environments.

However, DBR is also resource-intensive and challenging. Its iterative cycles can be time-consuming and unpredictable, sometimes exceeding available time or funding [9,10,11,23]. Researchers must also guard against confirmation bias arising from close participation in the design process, which can complicate objective analysis [10]. These challenges underscore the importance of structured, yet adaptive project management mechanisms that maintain DBR’s theoretical rigor while ensuring process accountability—an imperative that motivates its integration with agile development methodologies.

2.2. Agile Scrum for Educational Software Development

Agile development methodologies, and Scrum in particular, emerged in the software industry as a response to heavyweight, linear processes [14,24]. Scrum is built around a concise set of core principles that align closely with the rapid and iterative needs of immersive multimedia systems development. These principles are [25,26,27]:

• Iterative Development Cycles: Develop in short, fixed-length cycles (sprints) with a potentially shippable product increment at each sprint’s end. This allows for frequent delivery and testing of multimodal features.
• Prioritized Product Backlog: Features are represented as user stories that can be re-ordered or refined as needs change. The Product Owner represents stakeholders in prioritizing features, ensuring that resources are directed toward the most pedagogically and accessibly significant goals.
• Collaborative Team Structure: Daily stand-up meetings, sprint reviews, and retrospectives foster open communication and collective accountability. Defined roles—such as Scrum Master, Product Owner, and development team—create clear responsibilities for facilitation, prioritization, and execution.
• Adaptability and Responsiveness: If new requirements or user feedback emerge, they are incorporated into the next sprint rather than deferred to a long development cycle, promoting agility and inclusiveness in the design process.

In the context of educational software, Agile methods have been increasingly adopted to meet the evolving requirements of teachers and learners [14,28]. They provide the temporal flexibility and user-centered adaptability required when accessibility needs or pedagogical objectives shift during development [26]. For our purposes, Scrum’s insistence on continuous user feedback, incremental delivery and team transparency provided a built-in mechanism for iterative improvement that directly parallels the design-test-analyze cycles of DBR, making Scrum an operational complement to DBR’s epistemic orientation.

We recognize that common challenges of using Scrum in educational and accessibility-focused projects include scope drift, timeline misalignment between research and development, and dependency on highly cohesive interdisciplinary teams [29,30]. However, by proactively planning for these issues and leveraging the inherent flexibility of our hybrid approach, we were able to mitigate them, thereby enhancing Scrum’s responsiveness without compromising DBR’s reflective depth [31].

Together, these methodological foundations justify the integration of DBR and Scrum for accessibility-driven multimedia design. DBR contributes theoretical grounding and systematic inquiry, while Scrum contributes iterative cadence and development discipline—creating a hybrid model particularly suited to inclusive, user-centered innovation.

3. Related Work

Research at the intersection of DBR and Agile methodologies has expanded in recent years, yielding diverse approaches spanning educational innovation, methodological integration, and technology development. This growing corpus reveals how hybrid methodologies can enhance iterative learning design but also exposes persistent gaps in their application to accessibility-centered system development. Five representative contributions—Kastl and Romeike [32], Confrey [33], Fahd, Miah [19], Cochrane [34], and Smith-Nunes [4]—illustrate this trajectory and help position the distinctive contribution of the present study.

Kastl and Romeike [32] applied DBR and agile-inspired project structures to the secondary education context, focusing on how agile projects can foster cooperative learning in heterogeneous computer science classrooms. Their emphasis lies in managing learner diversity through project-based learning methods such as iterative planning, peer collaboration, and student-led user stories. While their study demonstrated the pedagogical value of agile classroom practices, it differs from the present work in both scope and epistemic orientation. Rather than addressing system design and accessibility challenges in educational technology, Kastl and Romeike [32] concentrated on learner engagement and differentiation at the instructional level. By contrast, the current study employs DBR and Scrum to address accessibility as a socio-technical design problem, developing an inclusive AR authoring tool for instructors and learners with diverse abilities in higher education.

Confrey [33] offered a reflective, autobiographical account of over three decades of DBR practice in developing learner-centered mathematics software. The integration of agile principles is framed as an organizational and leadership strategy for sustaining programmatic research, managing teams, and maintaining creativity across academic and corporate settings. The present work diverges from this focus on leadership and research program management by foregrounding methodological hybridization and accessibility outcomes. Specifically, the integration of DBR and Scrum here is directed toward accessibility-focused system design, producing empirically grounded best practices and challenges relevant to educational technology development rather than lessons in team organization.

Fahd, Miah [19] advanced the methodological discourse by proposing an integrated framework that combines DBR and Design Science Research (DSR) for developing a machine learning–based Educational Decision Support System (EDSS). Their framework emphasizes theoretical generalization through the alignment of DBR’s contextual inquiry with DSR’s artifact orientation. While sharing a commitment to methodological integration, Fahd, Miah [19] focus on bridging two research paradigms to support predictive model construction, whereas the present study blends DBR with a software development methodology (Agile Scrum) to yield actionable design practices for accessibility. Thus, their contribution is primarily theoretical and methodological, while this study is empirical, design-oriented, and inclusivity-driven, grounded in the practical development of an AR authoring tool.

Cochrane [34] introduced hybrid DBR for Agile Software Development (hDAS), a framework evolved through the design of Multi-User Virtual Environments (MUVEs) for vocational education and training. hDAS positions DBR as the driver of Agile tailoring, producing fit-for-purpose instructional design processes that evolve with theoretical and vocational demands. The paradigm shift proposed by Cochrane [34] significantly advanced instructional design research; however, the present study distinguishes itself by pursuing inclusivity as the primary methodological principle rather than as a derivative pedagogical concern. Rather than tailoring instructional design for vocational contexts, this project integrates DBR and Scrum in a balanced and reciprocal manner to address accessibility requirements in higher education technology design, treating inclusion not as a secondary outcome but as the explicit design objective.

Finally, Smith-Nunes [4] proposed AgileDBR as a generalized framework to bridge industry-driven Agile practices with academically grounded DBR. Her work emphasizes transparency, scalability, and ethical sensitivity, demonstrated through a case study on data ethics in postgraduate computing education. While both projects aim to integrate DBR and Agile, Smith-Nunes [4] provides a conceptual and prescriptive framework, whereas the present study delivers an empirical demonstration through the Accessible AR Authoring Tool Project. The distinction lies in orientation and epistemic contribution: AgileDBR formalizes a methodology for broad educational contexts, whereas the current work situates hybridization within the specific domain of inclusive multimedia system design, extending methodological insights into the design of accessibility-focused educational technologies.

Taken together, these studies demonstrate the richness of research at the intersection of DBR and Agile, while underscoring two persistent absences in the existing literature: (1) limited empirical evidence of hybrid frameworks applied to accessibility-oriented technology design, and (2) insufficient articulation of how hybrid cycles operationalize both research rigor and software-engineering efficiency. The present study addresses these gaps by presenting a concrete, accessibility-driven application of DBR–Scrum hybridization. Where prior work emphasizes either pedagogical practice [32], leadership and team management [33], theoretical integration [19], instructional-design adaptation [34], or framework generalization [4], this study uniquely contributes a design-based, accessibility-focused case that empirically validates how DBR and Scrum can be integrated to guide inclusive educational-technology creation. By situating accessibility at the methodological core rather than the periphery, this study reframes hybridization as a means of advancing both inclusive design practice and scholarly understanding of hybrid research–development processes.

4. A Hybrid DBR-Scrum Framework for Inclusive Multimedia Design

This study employed a hybrid methodological framework that integrates DBR for the overarching research design and evaluation with Scrum for software development and project management. This integration advances prior hybridizations by explicitly embedding accessibility as a central design principle throughout iterative cycles. While DBR provides an iterative, context-sensitive research paradigm for addressing authentic educational problems [6,7], Scrum offers adaptive, time-boxed cycles for managing software development complexity and ensuring responsiveness to evolving user needs [15]. The primary goal was to design and implement an open-source AR learning authoring tool that enables instructors with diverse abilities and technical backgrounds to create inclusive learning experiences. The framework operationalizes inclusivity through synchronized research and development activities, ensuring that accessibility validation occurs within each design iteration rather than only at the evaluation stage. The project was implemented in higher education contexts, providing a heterogeneous user base and ecological validity for the interventions.

The rationale for integrating DBR and Scrum lies in their shared commitment to iterative refinement, stakeholder engagement, and continuous feedback. Where previous studies (e.g., Confrey [33], Cochrane [34]) explored conceptual integrations of DBR and Agile, the present study operationalizes a structured hybrid model that synchronizes DBR’s research-oriented cycles with Scrum’s development-oriented iterations. DBR’s theoretical grounding ensures that design iterations contribute to educational knowledge, while Scrum’s empirical process control maintains development efficiency and stakeholder responsiveness [35]. The framework thus merges research rigor with engineering agility, embedding accessibility checkpoints in every sprint to ensure inclusion remains a formative, not terminal, concern.

Our methodological approach was structured around Plomp’s three high-level DBR phases [7]: preliminary research, prototyping, and assessment phases, which are dynamically synchronized with Scrum sprints (Figure 1). This integration depicts how DBR’s research-oriented cycles were operationalized through Scrum’s development-oriented iterations.

4.1. Phase 1: Preliminary Research & Analysis

The first phase focused on identifying the authentic design problem and establishing foundational requirements. In line with DBR’s principle of grounding design in real-world contexts [8], a literature review examined AR in education, AR authoring tools, accessibility evaluation of AR, and captioning strategies for immersive environments [2,36]. This was complemented by a document analysis of the Accessibility Guidelines, specifically the Web Content Accessibility Guidelines (WCAG) [37] and the Universal Design for Learning (UDL) principles [38], which provided normative and pedagogical benchmarks for inclusivity.

Empirical context analysis was then conducted through semi-structured interviews and surveys with potential users, including instructors with visual and hearing impairments [39]. This triangulated, multi-method approach aligns with DBR best practices that combine theoretical and contextual inquiry to frame design problems [21,40]. The analysis revealed a critical gap: existing AR authoring tools were largely inaccessible, lacking essential features such as synchronized captioning and multimodal navigation [39].

Drawing from these insights, a conceptual framework for the inclusive AR authoring tool was developed, grounded in UDL and WCAG standards. A product vision and an initial accessibility-prioritized product backlog were derived and expressed as user stories to maintain alignment with Scrum principles.

4.1.1. The Role of Accessibility-Focused User Stories

A cornerstone of our methodology was the early and sustained integration of accessibility-focused user stories. Prior work in inclusive design emphasizes the importance of embedding accessibility requirements from the outset rather than treating them as add-ons [41,42]. Accordingly, detailed user personas representing diverse users, such as a blind instructor, a deaf student, and a novice teacher, were developed based on preliminary research. These personas informed user stories written in the canonical format (“As a [user], I want [feature], so that [benefit]”) [43,44], which explicitly foregrounded accessibility needs.

For example:

• As an instructor with visual impairment, I want the interface to be fully voice-input navigable, so that I can create AR content independently.
• As a deaf student, I want synchronized captions for all audio in AR experiences, so that I can access equivalent information.
• As a novice teacher with limited technical experience, I want guided templates, so that I can author AR content without coding skills.

Each user story specified the user, the accessibility requirement, and its intended benefit, serving simultaneously as design guidance and acceptance criteria. By assigning high priority to these user stories within the product backlog, accessibility became a continuous driver of iteration rather than a final compliance task [45].

4.1.2. Tool Overview: Accessible AR Authoring Tool

To ground the hybrid methodological framework in a concrete artifact, this subsection provides an overview of the Accessible AR Authoring Tool developed through the DBR–Scrum process. The tool is intended to facilitate immersive, context-aware training and learning experiences within augmented- and mixed-reality environments. It enables subject-matter experts and educators to author and share holographic instructional workflows and digital learning content using wearable or mobile devices, thereby enhancing interactivity, situated learning, and practical skill transfer in workplace and higher-education contexts [46,47].

Key functional features include an intuitive, accessibility-centered authoring interface that allows users with limited technical expertise to create immersive learning scenarios; integration with 3D models and AI-assisted avatars; and the use of voice commands, visual feedback, and real-time expert guidance anchored in authentic settings. The platform supports multimodal interaction modalities—voice, gestures, visual cues, and haptic feedback—to accommodate the diverse needs of instructors and learners, including those with visual or hearing impairments, as represented in the accessibility-focused user stories [39].

Collectively, these features operationalize the project’s commitment to inclusive multimedia design by embedding accessibility not as a post-hoc enhancement, but as a core architectural principle throughout the authoring process.

4.2. Phase 2: Iterative Development & Formative Evaluation

The second phase comprised multiple DBR micro-cycles, executed through two-week Scrum sprints. The cross-functional team consisted of software developers (Scrum Development Team), educational researchers (one of whom acted as Product Owner to represent user needs), and a Scrum Master. In alignment with DBR’s collaborative ethos [5,48], practitioner collaborators, such as users with disabilities, were engaged as critical co-designers throughout.

During each sprint, developers implemented prioritized user stories, while daily stand-ups were adapted to incorporate both technical progress and reflections on user feedback. This ensured that research considerations informed development continuously. A kanban board was used to manage development and research tasks simultaneously, consistent with recommendations for hybrid Agile–research projects [4].

At the conclusion of each sprint, a Sprint Review presented a working increment to end-users for immediate evaluation. This aligns with DBR’s emphasis on testing conjectures in authentic contexts [49,50]. For example, an early prototype revealed a limitation in the voice-input navigation system when evaluated by a blind instructor; this feedback was directly incorporated into the subsequent sprint backlog. Sprint Retrospectives extended beyond Scrum’s process focus to also address DBR’s reflective stage, prompting the team to refine both research questions and data collection strategies.

Following each cluster of sprints, formative evaluations were conducted through usability surveys, observational studies, and semi-structured interviews. Findings informed the revision of design principles and the generation of new accessibility-focused user stories, thereby closing the DBR micro-cycle and reinforcing continuous improvement.

4.3. Phase 3: Summative Evaluation

The final phase involved a summative assessment of the intervention’s inclusivity and pedagogical effectiveness. In line with established DBR evaluation protocols [6,21], the finalized tool was deployed in a stakeholder workshop where instructors and students collaboratively created and explored AR learning materials. Data were collected on accessibility outcomes (e.g., WCAG compliance, usability ratings), learning outcomes (e.g., perceived ease of use, content comprehension), and user satisfaction [39,46].

This triangulated evidence provided a robust evaluation of the tool’s viability and its contribution to inclusive education. In contrast to previous DBR–Agile case studies that concluded at methodological reflection, e.g., Cochrane [34], this phase empirically demonstrates how hybrid cycles translate into measurable accessibility and usability gains.

All phases were systematically documented to ensure transparency and replicability, in accordance with methodological standards in DBR [21]. Collectively, the outcomes of these phases establish a replicable model for hybrid research and development cycles in accessibility-centered multimedia design. The findings from these evaluations are synthesized in the following sections as best practices and challenges for integrating DBR and Scrum in the design of inclusive multimedia systems.

5. Best Practices for Integrating DBR with Scrum in Inclusive Multimedia Design

The best practices presented in this section are empirically derived from observations, team reflections, and user feedback collected throughout the iterative DBR–Scrum development of the Accessible AR Authoring Tool. While the study does not employ inferential statistical analysis, it constitutes an empirical investigation in the DBR tradition, wherein evidence from real-world implementation and evaluation informs the derivation of methodological and design principles [5,21]. Accordingly, these practices represent analytical generalizations grounded in the documented experience of developing an accessibility-centered multimedia authoring tool, rather than speculative or conceptual propositions.

Drawing on the iterative development of the accessible AR learning authoring tool, this study identified eight empirically grounded best practices that facilitated the effective integration of DBR with Scrum. These practices, derived from observed team processes and design outcomes, extend prior methodological models by embedding accessibility as a continuous evaluative dimension within hybrid research and development cycles. They are proposed as guidelines for teams operating at the intersection of inclusive multimedia technology, user-centered design, and agile development.

5.1. Engage End-Users as Co-Designers

In DBR–Agile integration, end-users should be regarded as active co-designers rather than peripheral testers. From the earliest problem-analysis phase, we engaged instructors and students, including those with disabilities, through interviews, co-design workshops, and usability tests. Their involvement continued through sprint reviews and formal evaluations, ensuring that each development increment of the AR authoring tool was validated against authentic user needs. This practice aligns with DBR’s principle of collaborative partnership [5,6] and Agile’s user-centered ethos [51]. Prior research highlights that sustained practitioner participation enhances both validity and the relevance of design outcomes [8,52]. In our implementation, end-user feedback directly reshaped accessibility features—for example, refining voice-input navigation and caption synchronization after user testing—demonstrating the tangible value of participatory iteration. By embedding users in every sprint cycle, we not only surfaced issues early but also cultivated collective ownership of accessibility outcomes.

5.2. Use Accessibility-Focused User Stories

Accessibility requirements are often marginalized or treated as compliance checklists [53]. To counter this, we formulated and maintained user stories explicitly from the perspective of individuals with accessibility needs (See Section 4.1.1). This practice reinforced empathy and accountability within the development team, provided measurable acceptance criteria, and ensured accessibility remained central rather than peripheral to backlog prioritization. Inclusive design literature consistently recommends integrating disability personas and scenarios at the earliest stages of design [38,54,55]. Consistent with these principles, disability personas were embedded within our Scrum backlog and revisited in each retrospective to verify that accessibility tasks progressed in tandem with feature development. Our experience suggests that accessibility-focused stories function as “living artefacts”, evolving alongside user insights while continuously anchoring the team in inclusive design goals.

5.3. Align Iteration Cycles

A central methodological adaptation was synchronizing DBR’s iterative research phases with Scrum’s sprint cadence. One DBR cycle was structured around approximately two to three Scrum sprints, each culminating in a formative evaluation. This hybrid rhythm balanced Scrum’s rapid prototyping with DBR’s reflective inquiry [21]. Prior work has emphasized the value of aligning Agile iterations with DBR cycles to accelerate feedback and learning in educational contexts [33,34]. Our implementation demonstrates that aligning evaluation milestones with sprint reviews not only accelerates feedback loops but also enables accessibility metrics—such as compliance checks or usability heuristics—to inform theoretical insights in real time. Thus, each iteration became simultaneously a software refinement and a research inquiry cycle.

5.4. Define Hybrid Roles and Responsibilities

Blending research and development teams can create role ambiguity. We mitigated this through explicit hybrid role definitions. The Product Owner role was assigned to the lead researcher, ensuring accessibility remained a top priority in backlog management and that research questions were consistently addressed. Developers were encouraged not only to implement features but also to contribute observational data to research logs, while the Scrum Master coordinated sprint logistics to accommodate usability testing and data-collection needs. Confrey [33] emphasizes that clear role delineation is vital for sustaining DBR projects within Agile environments. Our findings suggest that positioning the researcher as Product Owner effectively balanced research and product imperatives, establishing a single point of accountability for accessibility compliance and theoretical contribution.

5.5. Maintain a Dual Focus: Product and Knowledge

A recurrent risk in DBR–Agile integration is privileging either the product (software artefact) or the research outcomes, to the detriment of the other [56]. We addressed this by treating both deliverables—functional prototypes and validated design principles—as inseparable outputs. Sprint reviews and retrospectives included reflective questions such as “Which accessibility principle was tested?” or “What design conjecture was confirmed or refuted?” Research tasks (e.g., “thematic analysis of user interviews”) were tracked alongside development tasks on the kanban board, ensuring parallel accountability. This approach resonates with calls to integrate practical and theoretical knowledge generation within design experiments [40,50]. By embedding analytical reflection into every sprint, the team maintained scientific rigor without sacrificing agile responsiveness.

5.6. Adapt Agile Artefacts for Research

Scrum artefacts were adapted to incorporate research-oriented deliverables. The product backlog included both feature stories and research stories, systematically tracking data-collection and analysis activities. Sprint backlogs and burndown charts were adapted flexibly, acknowledging that qualitative research tasks (e.g., coding interview transcripts) progress differently from programming tasks. We also maintained a design journal linked to backlog items, documenting how user feedback supported or challenged emerging design principles. This traceability is essential in DBR to link design decisions with empirical evidence [21]. Such modifications ensured that Agile artefacts served dual purposes: managing software development and maintaining methodological transparency across research phases.

5.7. Employ Scrumban for Adaptive Flexibility

While Scrum provided cadence and structure, we incorporated Kanban principles to create a Scrumban workflow that addressed the unpredictability of field research [57,58]. For instance, accessibility testing often depended on participants’ availability, necessitating real-time adjustments to sprint goals. By introducing work-in-progress limits and continuous-flow visualization, we accommodated emergent research tasks without compromising development discipline. Scrumban has been recognized as a pragmatic adaptation in complex contexts where both planned iterations and emergent tasks coexist [59]. In this project, Scrumban enabled flexible management of user engagement and data-collection cycles while retaining Scrum’s focus on iterative deliverables—a balance particularly beneficial for inclusive multimedia research.

5.8. Foster Continuous Communication and Reflective Practice

Finally, communication and reflection were prioritized beyond conventional Agile routines. In addition to daily stand-ups, we held mid-sprint checkpoints whenever user interactions occurred and biweekly consultations with accessibility experts. These mechanisms ensured alignment, minimized scope drift, and reinforced accountability to user needs. Agile scholarship consistently emphasizes communication as central to team effectiveness [60,61]. Our adaptation extended this principle to include cross-disciplinary dialogue between developers, educators, and users with disabilities, operationalizing DBR’s emphasis on reflection-in-action [21]. This sustained reflexivity enabled rapid adaptation and preserved methodological coherence across both research and development domains.

6. Challenges and Potential Mitigation Strategies

The integration of DBR and Scrum proved productive for developing the Accessible AR Authoring Tool, but introduced methodological and practical complexities not typically encountered in single-method approaches. This section discusses the empirically observed challenges and the strategies employed to address them, highlighting contextually grounded lessons for inclusive multimedia research and design practice. Each challenge and mitigation strategy emerged from documented observations, sprint retrospectives, and stakeholder reflections during the hybrid development process.

6.1. Balancing Research and Development Priorities

A persistent challenge was balancing the dual goals of producing a robust software artefact and generating empirically valid and theoretically informed research insights. Developers emphasized technical performance and timely feature implementation, while researchers required pauses for data collection, analysis, and reflective interpretation. This tension is widely observed in interdisciplinary projects where academic and industry priorities intersect [21,50]. In our project, this manifested when researchers required additional time for accessibility testing, delaying feature rollouts requested by developers.

Mitigation: We allocated explicit sprint capacity for both research and development tasks and alternated sprint emphases (“dev-heavy” vs. “research-heavy”) to manage competing priorities. Maintaining dual but synchronized roadmaps—for research and product development—ensured that milestones such as user studies aligned with relevant feature releases. The Product Owner role, held by the lead researcher, was instrumental in making informed trade-offs, echoing prior recommendations for unified leadership in hybrid projects [33]. This arrangement preserved research rigor without compromising software delivery velocity.

6.2. Scope Creep and Iteration Management

The openness of DBR and Agile created risks of scope expansion as stakeholder feedback revealed new possibilities. Such scope creep is a known challenge in design research, particularly where iterative discovery generates emergent requirements [56]. During tool development, additional accessibility requests—such as gesture-based navigation—surfaced mid-cycle, threatening to derail sprint plans.

Mitigation: We employed a “feature budget” per DBR cycle, prioritizing essential accessibility features and deferring enhancements to subsequent iterations. This strategy combined Agile’s Minimum Viable Product principle [62] with DBR’s concept of satisficing [56], ensuring the artefact remained functional for research purposes while preventing uncontrolled expansion. Documenting deferred features in a transparent backlog also maintained stakeholder trust and iterative focus.

6.3. Timing and Scheduling Conflicts

Scrum’s fixed sprint cadence occasionally conflicted with stakeholder and institutional schedules—for instance, academic holidays or examination periods that disrupted planned evaluations. Similar timing conflicts have been documented in other DBR projects [21,49]. our study, participant availability for accessibility testing often fluctuated, causing missed sprint reviews and delayed data collection

Mitigation: We adopted a Scrumban scheduling model, introducing buffer sprints to absorb delays and preserve overall cadence. This adaptation reflects Agile’s principle of “responding to change” [57] and has been recommended in hybrid Agile–research contexts [59]. The flexible Scrumban cycle enabled the team to maintain responsiveness to participants’ needs without undermining iterative momentum.

6.4. Ensuring Rigor in Data Collection and Analysis

Agile’s rapid iterations risked producing fragmented or anecdotal data, potentially undermining DBR’s methodological rigor. Similar risks have been noted in design experiments where reflection trails are not systematically maintained [50]. In our hybrid process, the pressure to deliver increments sometimes shortened reflection periods between sprints, risking incomplete data analysis.

Mitigation: We standardized data-collection instruments (e.g., identical usability and accessibility questionnaires across iterations) and introduced dedicated “analysis sprints” following major evaluations. During these sprints, developers focused on refactoring and technical-debt reduction while researchers analyzed qualitative and quantitative feedback. This dual focus preserved Agile momentum while maintaining research rigor. The use of shared analytic templates and consistent metrics enhanced comparability across DBR cycles.

6.5. Managing Iteration Fatigue

Repeated design and testing cycles led to participant and developer fatigue, with reduced enthusiasm for iterative evaluations by the third DBR cycle. Iteration fatigue is a well-documented issue in participatory design projects requiring prolonged engagement [63]. Instructors reported “feedback fatigue,” and developers expressed diminishing novelty in recurring accessibility tasks.

Mitigation: We maintained engagement by celebrating incremental successes during sprint reviews and sharing interim results at academic conferences. Publishing early outcomes provided external validation and motivation, aligning Agile’s ethos of celebrating small wins [64,65] with DBR’s emphasis on cumulative learning and impact [22]. These recognition mechanisms reinforced team morale and underscored the broader scholarly significance of continued participation.

6.6. Ensuring Transferability and Generalization

A recurrent critique of DBR is its limited generalizability due to contextual specificity [8]. In this study, concerns arose regarding whether findings derived from the Accessible AR Authoring Tool would extend beyond its immediate institutional context. Stakeholders questioned whether accessibility features designed for higher-education instructors could apply equally in vocational or informal learning settings.

Mitigation: To enhance transferability, we articulated underlying design principles (e.g., multimodal feedback, accessibility-first user stories, user co-design) alongside tool-specific insights. Prototypes were also tested in varied institutional contexts, including different faculties and disability-support centers. External expert reviews further validated general design principles, aligning with recommendations to document both process and principle for broader applicability [66,67]. This ensured that while the artefact was contextually grounded, the derived principles possessed cross-contextual relevance for inclusive multimedia design.

Collectively, these challenges and their mitigations illustrate that while integrating DBR and Scrum offers significant advantages for inclusive multimedia development, it also demands deliberate management of methodological balance, scope, scheduling, rigor, sustainability, and transferability. By systematically documenting these empirically derived challenges and their synthesis, this study contributes both practical guidance and theoretical validation for researchers and practitioners seeking to operationalize hybrid DBR–Agile frameworks in accessibility-centered educational-technology design.

7. Discussion

7.1. Synthesizing Findings: Linking Hybrid Processes to Inclusive Outcomes

Before examining the broader implications of these findings, it is essential to synthesize how the hybrid DBR–Scrum process directly impacted the final product outcomes. Continuous engagement with end-users through Scrum reviews and DBR reflections yielded tangible accessibility improvements—such as enhanced voice-input navigation, synchronized captioning, and multimodal feedback mechanisms. Scrum’s time-boxed cadence enabled measurable progress and accountability across sprints, while DBR’s reflective cycles ensured that every functional iteration remained pedagogically coherent and accessibility-aligned.

This synthesis illustrates that the hybrid methodology was not merely a procedural combination but a generative mechanism shaping inclusivity in the resulting AR Authoring Tool. It empirically substantiates DBR’s theoretical proposition that iterative inquiry in authentic contexts can coexist with Agile’s responsiveness to evolving user needs [21]. In doing so, it extends Agile scholarship on user-centered adaptability [34,59], providing concrete evidence that methodological convergence can bridge research rigor with practical design responsiveness.

7.2. Interpreting the Findings: Toward a Pragmatic Model for Inclusive Multimedia Design

This study demonstrates that integrating DBR with Agile Scrum constitutes a pragmatic and effective methodological framework for designing inclusive multimedia systems. The best practices identified—ranging from continuous user engagement to accessibility-focused user stories—function not as isolated interventions but as interdependent elements of a coherent hybrid model. Together, they address the dual demands of DBR (theory-driven inquiry in authentic contexts) and Scrum (empirical, incremental software development).

Previous research has highlighted tensions between rigorous knowledge production and efficient software delivery [68,69]. Our findings suggest that these tensions can be constructively managed when research activities are embedded within Scrum artifacts and ceremonies, ensuring that development remains evidence-driven while research outcomes remain practically grounded.

The model foregrounds accessibility as a central design principle, demonstrating that inclusivity need not be relegated to compliance checklists or peripheral testing [45,70]. By reconfiguring standard Agile artefacts—particularly user stories—as accessibility-first design instruments, the framework extends Agile’s user-centered ethos toward a more ethically grounded practice. This aligns with broader calls in HCI and multimedia research to embed equity and accessibility at the foundation of design processes rather than treating them as retrofits [54,71]. Hence, the resulting model operationalizes inclusivity not as a moral imperative alone, but as a methodological condition for valid and sustainable innovation.

7.3. Relationship to Prior Work

This work contributes to the growing literature on DBR–Agile integration by placing inclusivity at the methodological core. Previous studies have approached hybridization through various lenses: pedagogical practice [32], leadership and team management [33], vocational design adaptation [34], bridging research paradigms [19], and conceptual synthesis [4].

The present study complements and extends these works by providing an empirically substantiated account of how hybrid DBR–Agile processes directly yield accessibility outcomes in a functional multimedia system. It extends Confrey [33]’s claim that Agile enhances DBR sustainability by showing how Agile artefacts can be systematically adapted to prioritize accessibility. It also responds to Cochrane [34]’s concern about Agile’s tendency to privilege speed over rigor by demonstrating that synchronized DBR–Scrum cycles preserve both depth and delivery.

Unlike Kastl and Romeike [32], whose focus was on pedagogy, this work operates at the system-design level, providing actionable insights for HCI and multimedia researchers developing accessible digital tools. Similarly, while Fahd, Miah [19] and Smith-Nunes [4] propose theoretical integration frameworks, the current study provides an empirical complement, documenting concrete practices, challenges, and mitigation strategies observed during tool development. It therefore bridges the conceptual–empirical divide, demonstrating how methodological hybridization can evolve from theoretical construct to operational model for inclusive design.

7.4. Broader Implications and Future Work

The lessons derived from this project extend beyond the specific case of AR learning authoring. The practices identified—including accessibility-focused user stories, Agile ceremonies adapted for reflective inquiry, and Scrumban for iterative flexibility—constitute transferable strategies for other multimedia applications such as virtual-reality learning, cross-modal visualization, and interactive educational media for diverse users. These empirically validated strategies contribute to debates on inclusive design [72,73,74] by offering a replicable, ethically grounded framework for aligning technological innovation with societal inclusion.

Nevertheless, findings remain shaped by the specific institutional and technological context. As with most DBR studies, transferability relies on transparent documentation of both process and underlying design principles [12,66]. Future research should examine the scalability of this hybrid framework across diverse educational levels and media technologies, testing its adaptability to multi-institutional and cross-sectoral implementations. Longitudinal studies could also assess how hybrid DBR–Agile practices influence accessibility culture, team sustainability, and systemic change over time.

8. Conclusions

This paper presented a systematic and situated account of integrating DBR and Scrum to design inclusive multimedia technology. By strategically blending DBR’s iterative, theory-driven inquiry with Scrum’s structured, empirical development cycles, the study provides evidence that hybrid methodologies can produce functional systems while simultaneously generating transferable design knowledge.

The primary contribution lies in articulating a set of best practices and mitigation strategies—including accessibility-focused user stories, synchronized DBR–Scrum cycles, and adapted Agile artefacts—that directly support accessibility in immersive multimedia systems.

Theoretically, this study extends existing DBR and Agile scholarship by demonstrating how methodological convergence can sustain both research rigor and design agility. It empirically substantiates theoretical propositions that iterative cycles, when guided by inclusivity, can serve as engines of both innovation and ethical responsibility [21].

Methodologically, the hybrid framework offers a practical blueprint for synchronizing reflective DBR inquiry with the operational discipline of Scrum. It shows that research activities can be embedded within Agile artefacts—such as user stories, sprint reviews, and retrospectives—without compromising either methodological coherence or project velocity.

Practically, the findings equip educational-technology teams with actionable mechanisms—such as accessibility-first user stories, analysis sprints, and Scrumban scheduling—to manage inclusivity, scope, and rigor in real-world design projects. The proposed framework provides an evidence-based model that higher learning institutions, accessibility researchers, and multimedia developers can readily adapt.

Future research should evaluate the framework’s scalability and sustainability in diverse cultural and disciplinary settings, exploring its long-term effects on institutional accessibility practices and the development of inclusive design cultures. Comparative studies across other iterative design paradigms (e.g., Design Thinking, Lean UX) could further refine the theoretical understanding of methodological hybridity in inclusive system development.

Ultimately, this study underscores that the future of multimedia innovation must also be inclusive—ensuring that diverse educators, learners, and developers are not only beneficiaries but co-creators of technological progress. The hybrid DBR–Agile approach presented here provides a rigorous, adaptable pathway for achieving that vision—one that integrates scientific inquiry, engineering discipline, and social responsibility into a unified model of inclusive design.