Design principles are probably the most prevalent term used to characterise the theoretical understanding developed through iterative processes of educational design research. However, Gundersen (2021) finds in his review of 45 DBR studies that only one study reported on the development of guiding principles into conclusive or refined principles. Others did not mention design principles at all, or only put forward the guiding principles and then went on to conclude something else. Moreover, it is often not clear how specific design principles are articulated, who they address and how they are carried out in practice, despite attempts to clarify the term (van den Akker, 1999; Baumgartner & Bell, 2002; Bakker, 2019). In this way, it can be argued that DBR tends to favour idealised descriptions of design principles, rather than producing relevant and robust guidelines for educators and researchers to test and use in their relevant practises, which was initially one of the central ambitions of DBR during its infancy (Collins, Joseph, & Bielaczyc, 2004). This paper sets out to explore this lack of transparency and explicit articulation of design principles by focusing on what happens when design principles are applied in real-life teaching and learning settings. 

DBR researchers often rely on educators to carry out their principles in their practice to test their usefulness. Practises underly different possibilities and constraints. Through two empirical examples, we will show that the development and implementation of design principles often involve more “messy” or contingent processes of local negotiation than is suggested in most DBR literature. These negotiations may lead to unexpected findings beyond the aims of the originally planned design principles. Messiness is a well-known term within educational action research, where researchers engage in the messy practice of negotiating with practitioners on the contingent aims, methods and outcomes of the research process (Cook, 2009). Similarly, we regard messiness as an inherent aspect of any DBR intervention, even though there is a tendency in the DBR literature to present findings as the result of a linear refinement process towards a predefined goal. This means that we do not regard “mess” as a negative aspect of conducting DBR but as a basic condition when educational researchers explore design principles in “the swampy lowlands” (Schön, 1983, p. 42) of front-line practice. This points to a pragmatic need to analyse the possible impact and local adaptations of design principles and how the enactment of design principles always involves different forms of agency and responses from the involved social actors. Accordingly, in this paper, our research question is as follows: How can researchers articulate design principles in DBR and understand the agency achieved by educators and students when enacting design principles? We will address this question by first presenting a model for understanding key aspects of design principles and then describing two examples that show the messiness of enacting design principles in real educational contexts.

The importance of design principles has been emphasised by several DBR researchers, who suggest that they should serve to guide interventions and support the creation of new knowledge (van den Akker, 1999; Baumgartner & Bell, 2002; Gundersen, 2021). Design principles should be articulated in order to transform theoretical assumptions into specific guidelines that may address specific technological affordances, teachersʼ pedagogical approaches, or scaffold specific student tasks. In this way, design principles articulate intentions that can be applied and further developed through interventions in order to generate domain-specific knowledge.

Nieveen, McKenney and van den Akker (2006) suggest that DBR studies can be divided into either validation studies or development studies. Validation studies feature the design of learning trajectories to develop, validate and elaborate on theories about the process of learning and the implications for learning. In development studies, the fundamental aim is to develop design principles (van den Akker, 1999) for use in practice. Research is problem-driven, situated in the educational field and involves close interaction between practitioners, researchers, experts and other stakeholders. However, development studies may sometimes also be more explorative, as they seek to investigate new potentials, especially in the early stages of a research project (van den Akker, 1999). The guiding principles of an intervention can thus originate from either an existing theory (to be validated) or grow in the form of a problem or potential from an existing context (to be mitigated, explored or solved).

According to van den Akker (1999), it is important to discern between two aspects of a design principle. The first aspect is substantial – that is, the characteristics of the design itself. Substantive design principles can partly be extracted from prototype testing, which is one of the reasons why it is profitable early in the design process to analyse existing interventions to generate ideas for new design tasks (van den Akker, 1999). The second aspect is procedural – in other words, how the design should be developed or the characteristics of the design approach. Akker suggests that a combination of substantial and procedural principles may support design researchers in their tasks through heuristic statements in a format such as the following:

The intention of the statement is not to guarantee successful interventions, but to generate principles that allow depiction and discussion of the currently most appropriate knowledge for specific design and development tasks. 

Baumgartner and Bell (2002) also emphasise the relevance and articulation of design principles within DBR. The authors discuss the inadequacy of theories and explanatory frameworks in coming up with design solutions. Instead, researchers need to develop generative design principles, which should be informed by theories and, more importantly, help designers generate new designs to be tried out in practice. While some design principles can be generated after a design process is complete, the argument that these authors put forward is that in order to be useful, design principles should be accessible from the beginning of the research process (Baumgartner & Bell, 2002). As opposed to being explanatory in terms of giving reasons for an intervention being successful, useful design principles are generative in nature, leading designers to come up with new designs on the back of the guidance provided by the existing principles. This also involves creating criteria for evaluating usefulness (e.g. learning, goals, available class time, technical specifications, etc.), which can be used to assess the value of specific design principles.

To develop and explore design principles, Baumgarter and Bell (2002) suggest that researchers may benefit from addressing key questions, such as who is to apply the principles, what characteristics do the principles have, and how should the principle be applied. In addition to these questions, we wish to argue that design principles always need to reflect a why – that is, why is the principle carried out? This reiterates van den Akkerʼs (1999) emphasis on the specific purposes of arguments for applying design principles. These problems or potentials may be founded on different demands and theories put forward by different stakeholders, which could be internal (e.g. teachers or students) or external (e.g. funding agencies or educational designers) with a specific interest in validating the effectiveness of a given approach. It is, therefore, important that DBR manages to clarify what forms of knowledge or agendas influence specific design principles (cf. Buhl et al. (2022) in this special issue). 

Based on the discussion and categories above, we have created a model (shown in Figure 1), which illustrates how the articulation, enactment and assessment of design principles must address five key questions:

We propose that the model could be used as a heuristic tool for asking key questions when articulating, enacting, exploring and evaluating design principles during DBR processes. The questions in the model should be asked continually before, during and after iterative design and research processes. This could support researchers in navigating the contingency of DBR projects, which leads to the abandonment or revision of initial design principles, as well as the emergence of new design principles.

The two examples show that posing different questions with regard to the enactment of design principles in practice (i.e. the “how” (or procedural aspect), the “which” and the “why”) (see Figure 1) can be used to understand how design principles transform as a consequence of emerging design knowledge that diverges from the original design principles. The first example showed how a set of generic design principles was realised in a teaching unit facilitated by two teachers, where one of the teachers chose to ignore the pedagogical approach of the research project. This made the need for developing new design principles clear, which focused far more on the teachersʼ dialogic facilitation of the design-thinking activities. The second example showed how the preservice teachers questioned the practitioner–researchersʼ playful teaching methods, as it conflicted with their experience and knowledge of being students in teacher education. Based on this critical pushback, the practitioner‒researcher found it necessary to both revise the existing principles and develop new design principles. In this way, the examples show how the researchers in each project attempted to grapple with the messy response when enacting design principles in practice that sometimes supported and sometimes prevented the social actors from achieving agency through local design interventions. This illustrates how there is often a discrepancy between what is intended by a design principle and the response or pushback from practice. In this way, the aim of a DBR research process may be less about refining a design principle through a linear process and more about addressing this discrepancy (e.g. by identifying limitations, transforming assumptions or discovering new design principles).

The analysis of the two examples also suggests that different types of DBR (a validation study and an explorative development study) may involve substantially different possibilities for the involved participants (i.e. researchers, teachers and students) to enact and achieve agency. The validation study in Example 1 aimed to test a theory on teaching and learning (design thinking as a pedagogical approach) by trying out a design principle in its natural habitat to validate its effectiveness. In this way, the design principle was primarily enacted and defined through the agency that was enabled by a specific research agenda, as defined by the project researchers. The project showed how a teacher responded in an unexpected way. From an ecological perspective, the achievement of agency is informed by the teachersʼ personal and professional experiences, motivations for alternative future actions differing from the present and past, and the capacity to make practical and normative judgments and alternative actions in response to presently evolving situations. One could assume that teachers like Teacher P, who are able to draw upon a rich repertoire of experience, are more likely to share expansive orientations to the future and responses to present situations than might be the case (Priestley et al., 2015). Despite substantial capacity (skills and knowledge) and strong educational aspirations, innovation may be too difficult. Due to the fixed methodological procedure of the GBL21 project research design, there were limited possibilities for integrating the emerging knowledge on the need to support the teachersʼ dialogic facilitation in the large-scale project. By contrast, the explorative development study (Example 2) aimed to explore how problems and potentials in relation to playful learning emerged in a situated practice experienced by practitioners. Here, the aim was not only to test the theory behind the implemented design but also to improve the practice in which it was implemented, which could eventually be of value in other areas of teacher education. The development study, therefore, had a broader scope than the validation study, as it focused on the emergence of new knowledge though creative and experiential learning processes, and did not necessarily only focus on testing and contributing to knowledge regarding theories of learning. In this way, design principles in development studies may allow for more distributed agency when inviting a broader variety of local actors to engage in the negotiation of design principles.

We argue that the differences between the two research types also reflect a cultural and value-based discussion of teacher agency, which is often overlooked within DBR research. There is a long historical tradition in the Nordic countries and continental Europe for stressing the importance of achieving teacher agency in teaching, as well as in educational research (cf. Gundem & Hopmann, 2002). This bottom-up approach to teaching and research was clearly present in Example 2, where the practitioner‒researcher achieved considerable agency in terms of integrating the studentsʼ responses when adjusting and developing the design principles. The practitioner‒researcher is likely to achieve agency, as their position as both a developer and enactor of design principles implies social structures in the present moment that are not challenged by different and perhaps opposing experiences, values, beliefs, motivations and possible dilemmas in interpersonal relationships regarding, for example, power and trust. Such dilemmas would naturally apply, and be relevant when using the model as a reflective tool in other DBR projects where the social structures are different and, for example, are conditioned by a researcher who develops and enacts design principles in their colleaguesʼ profession. Unlike the practitioner‒researcher, the studentsʼ agency is limited. They are reflexive and seek plausible explanations for why the designʼs “how” and “which” do not match their expectations and previous experiences. However, as a result, the practitioner‒researcher articulates more specific design principles for the preservice teachers to focus on in the learning process instead of the curriculum. By contrast, the top-down approach of the GBL21 research project in Example 1 offered more limited options for teachersʼ agency. In this way, the implementation of design principles may involve important cultural and political tensions between what is intended by a research agenda and the possibilities of enacting agency at the local level of the classroom (Priestly et al., 2015).

Finally, we wish to acknowledge that working with design principles is challenging. The questions in the model may seem basic, as they are well-known within the research field of “didactics” (Jank & Meyer, 2002). However, as we have tried to illustrate, questions concerning design principles are in fact quite difficult to pursue in practice and, as pointed out by Gundersen (2021), actual descriptions of design principles and their transformations are often omitted in DBR publications. Formulating principles for practitioners to find useful, putting them into practice once understood theoretically, and identifying how to revise them based on experience gathered from real-life interventions, are all complicated tasks and yet lie at the heart of what DBR strives to achieve. The examples above show that researchers conducting DBR do not underestimate the agency involved when attempting to improve existing practices. The proposed model can scaffold the process of formulating principles, especially by stressing the importance of the procedural aspect, the need for identifiable criteria of success and reasons for relevance in a specific context. At the same time, the model can be used as an analytical tool when revising principles, where the five aspects can help focus the researchersʼ redesign process. By putting more emphasis on how design principles not only become refined but also abandoned or completely rejected, DBR researchers may be able to project a more realistic image of what it means to conduct DBR in response to messy real-life practice. Hopefully, this could contribute to not only more honest, but also more useful, educational research.

By proposing a model and analysing two empirical examples, we have identified possibilities and constraints when working with design principles in educational research and practice through DBR projects conducted in a Danish context. We argue that design principles should serve to guide the development and implementation of specific designs. This requires that researchers address the aspects of design principles presented in our model, as to why, who, what, how, when and which criteria apply to articulate, enact and evaluate the design principles. Furthermore, we argue that researchers should be able to articulate, enact and transform detailed design principles in iterative processes, and also be prepared for unexpected messiness that emerges when enacting design principles with respect for teachersʼ and studentsʼ agency. Thus, we argue that the continual articulation and adjustment of design principles should not only be seen as refinement, but also as a contingent process involving local agency, allowing unexpected challenges and valuable insights to emerge.