Educational design research: Portraying, conducting, and enhancing productive scholarship

Abstract Context Solutionism is the all‐too‐common human propensity to jump to a solution before adequately understanding the nature of a problem. Solutionism has long been prevalent in efforts to improve education at all levels, including medical education. Thesis Educational design research (EDR) is a genre of research that features the gaining of in‐depth understanding of a problem before any prototype solution is designed and tested. It is different from other forms of scientific inquiry because it is committed to the simultaneous development of both theoretical insights and practical solutions, together with stakeholders. This approach is powerful for theory building because it privileges ecologically valid studies that embrace the complexity of investigating learning in authentic (as opposed to laboratory) settings. When conducted well, both the research process and its outcomes generate valuable contributions to practice. Preview This article constitutes an expository essay on EDR, comprised of three movements. First, the approach is defined, its origins are presented, and its characteristics are described. Second, a generic model for conducting EDR is offered, and illustrated with examples from the field of medical education. Third, pathways towards advancing this form of inquiry are discussed, including ways to address inherent challenges and limitations, as well as recommendations for the medical education community. Although EDR is no panacea, this article illustrates how it can serve medical education research in a wide variety of geographic and disciplinary contexts.


| 83
McKENNEY aNd REEVES A distinction has often been made between basic research to discover new knowledge and applied research to solve practical problems, but this simplistic dichotomy does not adequately represent either how research is actually conducted or the multiple goals pursued by most scholars. 5 Education researchers, including those working in the field of medical education, often have a range of different goals in that their purposes may be descriptive, predictive, interpretivist, or refer to development or action. 4 For researchers interested in contributing to theory alongside development goals, educational design research (EDR) may be a compelling option. 6 We observe that the achieving of complex development goals is rarely feasible through simple, linear or predictable pathways, and this is also the case in medical education.

| P ORTR AYING EDR
'Educational design research can be defined as a genre of research in which the iterative development of solutions to practical and complex educational problems also provides the context for empirical investigation, which yields theoretical understanding that can inform the work of others.' 7 Identifying problems amenable to EDR involves finding real-world challenges that are worthy of investigation and capable of being solved through the EDR process. Specific problems may be identified by practitioners, by researchers, or through the study of literature. 7 For example, a serious problem in medical education refers to helping future physicians develop consistent habits to prevent sepsis. 8 Over a million cases of sepsis occur in United States hospitals every year and 15%-30% of them result in death. 9 Although antisepsis protocols are well known, research is needed to understand why they are insufficiently adhered to, and to develop additional solutions that can eradicate this problem. 10 When even an obvious serious problem is raised, verification in both literature and practice is necessary to ascertain if it is, indeed, legitimate, researchable and research-worthy. From the theoretical perspective, the problem is worth studying if doing so would address a clear gap in the existing literature (legitimate), if existing methods will allow it to be studied well enough to warrant the effort (researchable), and if the work will contribute to theory development or scientific understanding related to a widely held, as opposed to idiosyncratic, concern (research-worthy). 7 From the practical perspective, the problem is worth solving if the real problem, as opposed to a symptom, is identified (legitimate), if it can be identified in accessible contexts (researchable), and if it is severe enough to encourage stakeholders to invest in solving it (research-worthy). 7 Here is an example situated in a gross anatomy class: • Problem: corpse donors are limited and existing simulations for teaching human anatomy lack sufficient fidelity (the practical side of the problem). The current simulation literature provides insufficient guidance on how to develop high-quality simulations for mammalian anatomy (the scientific side of the problem).
• Practical aim: to develop a high-quality human dissection simulation that allows students to meet course goals without conducting actual dissection.
• Scientific aim: to understand and describe the characteristics of high-quality simulations for mammals in general and humans in particular.
Research that does not explicitly seek to contribute to both theory and practice by addressing real-world challenges can certainly be of great value, but it does not constitute EDR.
Undertaken in three movements, the purpose of this paper is to introduce EDR to the medical education community. The remainder of this section further portrays (the origins of) the approach, as well as similar approaches found in medical education research. The second section of this article describes how EDR is conducted and gives examples from the field. The third section considers productive pathways forward, in light of inherent challenges and limitations.

| Theoretical and practical synergies: a brief, non-comprehensive review
The notion of a linking science connecting theoretical and practical work has been advocated by psychologists for over a century. 11,12 In the 1930s and 1940s, major advancements in this direction were made by Lewin and colleagues, through action research, in which hypothesis generation and testing through the discussion of problems followed by group decisions were central. 13 In the 1960s and 1970s, (participatory) action research flourished in the social sciences, practitioner inquiry emerged, and calls for educational research to directly address the problems and needs of education increased. 14 Amongst other things, this set the stage for the rise of 'action science' in the 1980s 15 and the notion of 'use-inspired basic research,' which gained widespread momentum in the 1990s 5 . Around that time, researchers in the fields of instruction design and curriculum development began to stress the need for more reliable, prescriptive understanding to guide the robust design of educational products, programmes, processes and policies. [16][17][18] At the same time, researchers in the field of education psychology published landmark papers arguing for how theory informs the design of learning and vice versa, calling for research to be situated in the contexts in which that learning actually takes place, and citing the shortcomings of laboratory settings for understanding learning phenomena. 19 for more research like that of the French chemist and microbiologist Louis Pasteur, who sought fundamental knowledge within the context of solving real-world problems such as the spoilage of milk and treatment for rabies. In this tradition, EDR is concerned with the solving of existing problems in practice and with the structuring of the inquiry process so that it yields scientific understanding that is ecologically valid and informs the work of others.

| A family of approaches
We use the term 'EDR' to describe a family of approaches that strive towards the dual goals of developing theoretical understanding and also designing and implementing interventions in practice. 21

| Examples of EDR in medical education
Like other research, EDR extends existing theoretical knowledge through data collection and analysis. [23][24][25] However, unlike many other kinds of research, the EDR process is embedded in the (often cyclic) development of a solution to the problem being tackled. [23][24][25] Here are three examples: Although brief, these descriptions show how each of these studies developed theoretical understanding and contributed to the improving of practice through the design and testing of interventions.

| Key characteristics
In the same way that engineering design melds creative insights with pragmatic understanding and follows the best available theoretical principles derived from physics, materials science, aesthetics and other disciplines, EDR is a complex and multifaceted endeavour. The simultaneous pursuit of practical and scientific goals is central to the process, which can be characterised by five essential features. Educational design research is 'theorectically oriented' not only because it uses theory to ground design, but also because the design and development work is undertaken to contribute to broader scientific understanding.
It is 'interventionist' because it is undertaken to engender productive change in a particular education context. It is 'collaborative' because it requires the expertise of multidisciplinary partnerships, including researchers and practitioners, but also often others (eg, subject matter specialists, software programmers and facilitators). It is 'responsively grounded' because its products are shaped by participant expertise, the literature and, especially, field testing. Finally, it is 'iterative' because it generally evolves through multiple cycles of design, development, testing and revision. Given these characteristics, it will come as TA B L E 1 Family of approaches seeking practical and scientific synergies, including those that inherently yield theoretical and practical outcomes through intervention development (grey cells), and those that may yield theoretical and practical outcomes, depending on how they are used (white cells)

Approach Goals Key characteristics
Design-based research (DBR) 6,44,49 To enhance understanding about the nature of learning and what facilitates it • Takes place in continuous cycles of design, evaluation and redesign • Takes place in authentic real-life learning settings in which learning occurs normally • Is aimed at both testing and refining theories, and advancing practice • Is characterised by mixed-methods studies • Involves designers, researchers and practitioners with different expertise who interact frequently to guide the design, conduct and reporting of DBR To critique and deconstruct power inequities in society with the design of practical solutions to serious problems underlying such inequities • Advances fundamental insights about human learning and development • Exposes explicit or implicit normative hierarchically powered decision-making structures and related assumptions of objectivity • Critically attends to a range of theoretical lenses (eg, colonial, racialised, gendered, queered) during design and partnering

Medical Research
To employ experimental methods as well as dealing with the complexity inherent in public health and education innovations • Develops evidence base, theory, models for processes and outcomes • Pilots with attention to feasibility, recruitment and retention, sampling • Evaluates to assess effectiveness, understand change processes, assess cost-effectiveness • Implements with attention to dissemination, monitoring and follow-up Action research 58,59 To address problematic situations in organisations or communities

| Modelling the process
The present authors 7 previously surveyed models for EDR, as well as for instruction design and curriculum development (eg, Ejersbo et al, 29 Bannan-Ritland, 30  Although additional information is given in the source publication, 7 three main features bear mention here. First, the squares in Figure 1 denote three core phases, and the arrows between them indicate that the process is both iterative and flexible. During the 'analysis and exploration' phase, collaboration with practitioners is sought in order to shape a better understanding of the problem to be addressed. Research during this phase is typically informed by and contributes to theoretical understanding concerning the problem, context or stakeholders. During 'design and construction,' ideas about how the problem might be addressed tend to start off as rather large and vague, and gradually become refined, pruned and operationalised. Although influences the ongoing core processes, as well as the ultimate outputs, and vice versa. Although not shown here, the professional development of those participating in the study (practitioners and researchers alike) is often a by-product of the overall process and especially of the implementation work.

| One example spanning all phases
To illustrate how this model comes to life, we briefly describe a previously published 4-year PhD study, 32

| Multiple examples highlighting contributions from different phases
Across the shared characteristics of EDR, differences are also pre-  Table 2.

| Challenges and limitations
Phillips and Dolle, 42   Every research approach has its limitations and EDR is no different in this regard. For those considering EDR, it is important to make informed choices. First, EDR requires close collaboration between (at least) researchers and practitioners. 43 When this is not feasible or desirable, EDR ceases to be a viable option. Second, EDR connects theory, innovation and practice. 44 When the development of one or more of these is not of high priority, EDR is not likely to be useful.
Third, because it centres on creating productive change in practice, EDR requires substantial amounts of time. 45 If the time available is measured in weeks (rather than months or years), EDR is not likely to be feasible.
Finally, given that EDR is fairly easy to do poorly and quite difficult to do well, a fourth limitation of EDR has less to do with the approach per se and more to do with the capacity of those conducting it. As with other ideas, the value of EDR lies not in its definition but in Develop competence with a wide range of qualitative and quantitative methods, to enable selection based on purpose As with other forms of inquiry, educational design research (EDR) is easy to do poorly Clearly meet the standards of rigour associated with the (qualitative, quantitative or mixed) methods being used

Communicative
The study seems too large or too complex to report in one (eg, article-sized) chunk Portray design projects as a collection of sub-studies, reported separately, each making a significant contribution in its own right, and remain mindful that unnecessary discussion of the overall study can be distracting to the audience Because specific contexts are involved, the value to others (outside the research setting) seems limited Situate the work as a (multiple) case study, clarifying the nature of the case and the frequency with which this phenomenon occurs, as well as also specifying salient details so that readers can ascertain if case-to-case generalisation is applicable Political Organisational barriers (eg, disciplinary barriers, human capacity) inhibit the work Identify and focus work within the jurisdiction of change represented by the areas(s) within which the design project is able (by its own authority or through influence) to decide upon and implement change The power or opinions of stakeholders are valued or used in conflicting ways Where possible, use facilitation skills and ethical guidelines to help parties negotiate priorities. In so doing, help all to see that deep and lasting change especially requires the commitment of those who will directly implement and sustain it (analysis and exploration; design and construction; evaluation and reflection; implementation and spread). They articulate four foundational competencies that are required to enact each role, and offer guidelines for developing them. They note that 'empathy' is needed, for example, to explore (un)shared goals or become exposed to the incentives, motives and reward structures in different settings.
'Orchestration' helps to simultaneously attend to research framing, data collection, solution design, implementation, infrastructure woes and stakeholder ownership. Creative and analytical 'flexibility' supports the optimisation of the human and material resources available in ways that remain aligned with the goals of the instruction. Finally, 'social competence' refers to a robust repertoire of the interaction strategies needed to fulfil each role. Thus, conducting EDR is clearly a complex task. If this form of inquiry is to realise its potential contribution to the field of medical education, explicit attention must be given to the holistic development of design researcher capacity.
As is the case with other forms of complex learning, this requires engaging with entire authentic design research projects. In deciding whether EDR is an appropriate fit for a given project, scholars are advised to consider the substantial and varied demands placed on researchers undertaking this kind of work.

| CONCLUSIONS
Educational design research is of course no panacea. However, it does put the metaphorical brakes on solutionism because of its heightened attention to clarifying the nature of the problem before an educational intervention or solution is conceived. We believe that medical education faces many challenges that might be best addressed by synergistically pursuing both theoretical and development goals. For example, as Chen and Reeves 48 argue, this approach could be used to: • develop capacities to work effectively in increasingly fluid health care teams; • cultivate skills to communicate in a culturally competent manner with patients and other health care professionals; • prepare health care professionals for practice in a world that is increasingly infused with machine learning algorithms and robots; • improve assessment protocols and feedback practices to promote competency-based education, and • enhance health care professionals' clinical reasoning skills.
This article set out to discuss the nature and origins of EDR, how it is conducted, and what is needed to advance this kind of work. First, the approach was introduced, discussed in the light of related developments in the last century, as well as other contemporary approaches that strive towards both practical and theoretical goals, and characterised. Second, a generic model for conducting EDR was described and illustrated with multiple examples. Third, the challenges and limitations of this approach were considered. We conclude this article with a few words about next steps.
According to its website, this journal promises its readers 'practically oriented and theory-informed papers that emphasise empirical evidence and advancing the field.' As such, it would seem that EDR's twin pursuits align well with the ambitions of Medical Education. If fluency with the approach has yet to be developed before this community shares its design studies (more), then collaboration with researchers in sister fields, who are more accustomed to design research (eg, those in the learning sciences, instructional design or educational technology) may be worth exploring. In so doing, it seems crucial to seek out like-minded scholars who prioritise the giving of careful attention to ensuring descriptive and explanatory understanding of problems worth tackling before developing solutions.
The contributions to this special issue on solutionism offer multiple starting points for doing just that.

AUTH O R CO NTR I B UTI O N S
SMcK and TCR have been collaborating on the articulation of this approach for over a decade. This piece is based on that collaborative effort. SMcK drafted, revised and finalised the present text and figures. TCR edited and revised the present text. Both authors (SMck and TCR) developed the core ideas described in this paper, approved the final manuscript for submission, and agreed to be accountable for the work.

ACK N OWLED G EM ENTS
The authors would like to thank Professor Diana Dolmans (Faculty of Health Medicine and Life Sciences, Maastricht University, the Netherlands) for encouraging this submission.

CO N FLI C T S O F I NTE R E S T
Not applicable.

E TH I C A L A PPROVA L
Not applicable.