Applying a conceptual design framework to study teachers’ use of educational technology

More than a decade ago, Sandoval (Sandoval 2004) commented on the need for theory development in DBR. He argued that a theoretical framework was necessary in order to more systematically study educational design in different contexts and to develop a more general theoretical knowledge that would inform educational design in other than localised contexts. Gravemeijer & Cobb make a distinction between two different approaches to the development of theories in DBR: the development of local instruction theories and the development of ‘theoretical frameworks that address more encompassing issues’ which could be used at a higher abstract level to guide teachers’ design for learning (Gravemeijer and Cobb 2006, p. 19). However, more recent research indicates that DBR has mainly focused on developing local instruction theories in the form of design principles as outcomes of interventions to meet local design challenges (McKenney and Reeves 2012, p.36). Anderson and Shattuck (2012) conclude that although DBR is a promising research approach, its aim to provide theoretically based generalisable advice for educational designers has not yet been realised. Holmberg (2014) suggests shifting the focus of attention in DBR from the practical intervention and its characteristics to the teacher’s cognitive processes and actions before, during and after the design process. A shift in focus from design product to design process and teachers’ design thinking would also mean a change in the theoretical contributions of DBR. A theory-based and empirically tested design framework could provide researchers with a tool to study teachers’ design with ICT in different situated contexts. Instead of being restricted to what has worked in a certain context for a certain problem, a conceptual design framework could also help to support teachers’ design thinking at a more general level. Conceptual frameworks are context independent in the sense that they retain their basic “meaning” regardless of the context. However, the unique context in which they are applied also affects how they are enacted in practice. In education, Biggs’ idea of ‘constructive alignment’ is an example of a conceptual framework that underlines the importance of aligning curricular objectives, the activities performed by the teacher and students and the teachers’ assessment of students’ learning outcomes (Biggs and Tang 2011). Although the idea of alignment as a ‘framework’ can be applied to different educational settings, the formulation of specific learning outcomes and the choice of educational activities will always be adapted to the unique design context. A design framework would aim to represent the roles of the teacher, the individual student and peer learners in their individual and collaborative meaning making. Gravemeijer & Kobb suggest that the complexity of educational contexts imply the need for a broad design framework in DBR and one that incorporates ‘…general background theories such as socioconstructivism, or sociocultural theory, domain-specific theory and theories on specific elements of the learning ecology…’ (Gravemeijer and Cobb 2006, p. 48).

Research indicates that a successful educational design is one in which the learner’s preconceptions are elicited and where the learner is encouraged to build on, modify or refute these preconceptions (Bransford et al. 2000). Successful learners develop subject specific conceptual frameworks that allow them to understand, select and critically interpret new information (Biggs and Tang 2011). This insight is closely connected to research findings that stress the importance of learners developing metacognitive strategies and teachers creating designs that help students to develop such strategies (Bransford et al. 2000). There is also evidence to suggest that learners benefit from multimodal examples and variations of the concepts, ideas and skills to be learned, preferably in different contexts (Hattie and Yates 2014; Marton and Pang 2006). These research based insights are useful for teachers considering using ICT to support every learner’s cognitive process. However, social constructivist and sociocultural theories of learning, and more recently connectivist perspectives of learning, emphasise the social aspects of learning (Rogoff 1995; Siemens 2005; Weick 1995; Wertsch 1991; Wenger 1998; Vygotsky 1986; Zittoun and Brinkmann 2012). In line with this research, in this article learning is discussed in terms of meaning making:

“Learning as meaning making” is an expression emphasizing the fact that in any situation of learning, people are actively engaged in making sense of the situation – the frame, objects, relationships – drawing on their history of similar situations and on available cultural resources. (Zittoun and Brinkmann 2012, p. 1809)

Viewing learning as meaning making means that a central concern for teachers is how to stimulate and support an exchange of knowledge representations, and communications about these, between teachers and student(s) and between students. A theoretical added value of ICT is its potential to stimulate and support this exchange in new ways (Sawyer 2014; Spector et al. 2014). Thus, a design framework could be used as a model or an analysis tool when looking at or thinking about educational designs.

Vygotsky illustrates how learning through communication is different from learning through instruction and practice. When learners communicate and express ideas, they contribute to what it means to know these ideas and how this can be expressed in speech. This illustrates how thought ‘does not merely find expression in speech; it finds its reality and form’ (Vygotsky 1986, p. 219). Similarly, Weick emphasises that learning is about making meaning that materialises through communication (with others and oneself) and ‘about authoring as well as interpretation, creation as well as discovery’ (Weick 1995, p. 8). Papert (1980) maintains that an especially powerful way of learning through interaction and communication is individual and collaborative learning-by-making. This process includes an externalisation of ideas, which makes them tangible and allows individuals to ‘see their thinking’. When we act on the world together with others we also make our ideas and knowledge shareable and communicable. Drawing on the research discussed above, it is possible to argue that a central part of a teacher’s work is to support meaning making by stimulating and supporting the exchange of and communication about abstract conceptualisations in the form of knowledge representations between teachers and students and between students. By using ICT, the process of meaning making can theoretically be supported across the boundaries of time and space and through different modalities (Sawyer 2014; Spector et al. 2014; Stahl et al. 2014). However, as indicated at the beginning of this article, previous research indicates that teachers could benefit from support in this. A number of theoretically informed conceptual models and frameworks with various foci, complexities and levels of abstraction are available to help us understand the different aspects of educational design with technology.² The Conversational Framework (CF) was developed by Diana Laurillard with the intention of supporting teachers in their design with ICT (Laurillard 2002, 2012). IT has been applied in various learning settings and subject areas (Sharples et al. 2007). In line with the above discussion about learning as meaning making, the CF recognises the interdependence of individual and social processes in learning and that an educational design should reflect this by creating conditions for communication and the exchange of knowledge representations between individual learners, teachers and peers (Laurillard 2012). The CF represents the roles played by these actors in formal learning contexts and can thus provide a conceptual framework for how teachers, through the use of ICT, could design a teaching-learning environment for multimodal ‘conversations’ between them.

In Laurillard’s model, the individual learner is placed at the centre (see Fig. 1). The grey area represents what Laurillard refers to as the learner’s internal learning cycle. Laurillard describes how the teacher’s role is to ‘motivate the internal cycles generating and modulating the learner’s concepts and practice, which is what facilitates learning’ (Laurillard 2012, p. 86). An individual learner can interact with other actors by means of five communicative cycles at two levels: the conceptual level and the practice level. At the conceptual level, the students’ meaning making processes are supported by communication about what is to be learned (Laurillard, p.87). An example of learning supported at the conceptual level is when a teacher explains something orally and the individual student or peer asks a question or generates an articulation of his/her own conceptualisation. The teacher can then answer the question and give feedback on the articulations. This is known as the teacher’s communication cycle (TCC). When the individual learner and peers exchange articulations of their concepts, for example when discussing how to interpret a text or solve a problem, this takes place in what Laurillard refers to as the peer communication cycle (PCC).

However, a learner’s conception of what is to be learned can also be expressed or supported at the ‘practice level’ (Laurillard 2012, p. 89). In formal learning contexts, ICT can be used to design what Laurillard refers to as a teacher’s practice/modelling environment (TPME). Here, students can participate in a number of (learning) activities, primarily with their teacher and peers, although the use of ICT increasingly permits interactions with external actors. In the TPME, the individual learner and peers act on the external environment by applying concepts. Thus, the individual learner’s practice and the peers’ practice are represented in the TPME as representations indicative of the students’ understanding. The teacher can give feedback in relation to these representations in the teacher practice cycle (TPC).³ The TPC can thus help the learner ‘connect’ a representation at the practice level with the concept that is to be learned, thus supporting the learner’s internal learning cycle. The TPC also provides teachers with feedback on their teaching and whether students are learning what is intended.

The author studied the design practices of eight experienced upper secondary teachers of English as a foreign language (EFL) in four Swedish municipal schools with one-to-one laptop programmes. Collaboration with as many as eight teachers could be considered atypical in DBR. However, when examining the usefulness of a more generic conceptual construct (i.e. the CF) as an analytical design framework, applying it as a conceptual lens in different design contexts was considered important (Barab 2004). Empirical data was collected over a period of fifteen weeks, starting in December 2014. The teachers had all taught English for more than six years and had volunteered to participate in the project after they and their colleagues had received information about it. They all had a positive attitude as far as the pedagogical potential of ICT was concerned and considered themselves somewhat more ‘tech savvy’ than the average colleague. Nevertheless, all but two of the teachers claimed that their technological knowledge was limited and needed to be substantially developed. This was the first DBR project they had participated in. During the study the author met each of the teachers three times. Other means of communication were used between these sessions and are described in the data collection section below. Together with each teacher, the author decided which of the teacher’s classes to focus on to follow his or her authentic design practice.⁴

A second source of data is the teachers’ educational designs that they created and used in their digital teaching environments, and to which the researcher had full access. A third source of data is the reflective logs that the teachers were asked to keep and share with the researcher between the design conversations. Thirty-one log entries were received, varying in length between some 60–250 words. A fourth source of data is the researcher’s own field notes. The triangulation of data sources described above enabled the researcher to capture different dimensions of the teachers’ practices and arguably contributed to the trustworthiness of the analysis process (Flick 2014; Miles and Huberman 1994).

In the first part of the analysis process the collected data is coded and analysed according to the principles of what is referred to as a directed (Hsieh and Shannon 2005), or deductive (Elo and Kyngäs 2008) approach to qualitative content analysis. This approach is often used when the goal is to validate or conceptually extend a theoretical framework, model or theory (Elo and Kyngäs 2008, p. 111; Hsieh and Shannon 2005, p. 1281). The CF is used to inform a coding frame referring to its five different communicative cycles. An additional “Other” category is included too. In the event that data is considered to relate to an idea about or an example of technology-supported design to support learning, but cannot not be coded using the five categories in the matrix, it is coded “Other”.

During the reading of the transcribed design conversations and the teachers’ written reflective logs, the researcher selected, but did not code, the parts of the text relating to the educational use of technology. Here, the intention was to not use the CF as a conceptual lens. This was done in an attempt to increase the trustworthiness of the analysis by making the researcher ‘see’ other uses of technology to support learning that corresponded with the communicative cycles in the CF. In the next step of the analysis, the selected passages were coded using the six main categories described above. It was decided that if the “Other” category was used extensively during the coding it would indicate that the initial coding frame needed to be revised. During the coding the researcher read the selected passages in the transcriptions while simultaneously listening to the original recording of the passages. This was done in order to pick up on nuances, pitch, hesitations and so on that were not reflected in the written transcript but that could be important when interpreting and understanding the data. The same procedure was used two weeks later, when the passages were double-coded by the researcher to increase intracoder reliability. In the next step of the analysis the researcher read through all the coded examples again, inductively coding and categorising the data segments with the intention of identifying salient features of use in the main categories and sub-categorising them. This step of the analysis was repeated a second time two weeks later.