The purpose of this chapter is to highlight the current choice of methodologies that are preferred in artificial intelligence in education (AIed) studies and propose a value-added methodology for such studies. After a systematic literature review of the methodologies deployed in AIed studies, it can be seen that object-oriented methodology (OOM), action research (AR), and grounded theory are most often deployed. However, there is a case to be made for design-based research (DBR) methodology. DBR has a natural inclination with AIed (given their common history), is not bound by ontological presuppositions, and has the ability to generate knowledge and theory grounded in practice. These reasons alone show the value-addedness of DBR. A further comparison against the currently methodologies deployed will show that DBR has the ability to compensate for each of the weaknesses of the other methodologies.
TopBackground
A quick view of the methodologies used in the building of AIed frameworks and applications show that the vast majority of studies do not use a methodology per se, or use a weakly applied methodology (Qin et al., 2020). This is surprising given the rate of pace of development of AIed (Chen et al., 2020).
Design based research (DBR) is a research approach that tries to solve complex practical problems by utilising ‘design principles’. DBR has its roots in the field of education and still remains highly popular in educational research (Anderson & Shattuck, 2012; Gamage et al., 2016). AIed is the application of artificial intelligence in education, which resides also in the fields of artificial intelligence and education. Like the thrust of most AIed applications, DBR focuses on the solving of real-world problems and tends to be grounded in real world experience framed within theoretical models. By applying theoretical design and involving various stakeholders, a theoretically sound foundation tempered by real world experience gives a well-designed solution.
However, despite the seemingly natural compatibilities of applying DBR in AIed research, DBR has been unusually seldom used (Zhang & Aslan, 2021). This is very surprising due to the fact that both are in the same field, with the same leanings and would be very useful to explore if there is a case to be made for DBR to be applied in AIed.
In order to gain deeper insights into and investigate Design-Based Research (DBR) within the context of Artificial Intelligence in Education (AIed), the study aims to answer the following research questions:
RQ1: What are the more commonly used methodologies in AIed?
RQ2: What is the case for DBR methodologies in AIed research?
RQ3: How can DBR be adopted for the development of AIed frameworks?
TopResearch Philosophy
Ontology is defined by the APA Dictionary of Psychology as “the branch of philosophy that deals with the question of existence itself” (APA, n.d.). In layman terms, this examines the nature of reality. Building further to this, is the concept of epistemology, which is defined as “the theory of knowledge embedded in the theoretical perspective and thereby in the methodology” (Crotty, 2020). Together, these two concepts form parts of what is often understood as a ‘paradigm’. In social research studies, “paradigm” can be understood to refer to the ‘philosophical assumptions or to the basic set of beliefs” that inform the worldview and guide the resulting actions (Lincoln et al., 2017). One such paradigm includes pragmatism.