Abstract
This paper reflects on a popular and influential theory unique to information systems (IS), namely representation theory, which is widespread in research on conceptual modelling. We review scholarly work in this domain and discuss why and how studies using representation theory need to transcend their focus of research and link their findings to further consequential variables of interest. We propose an innovative research design that builds upon, and converges, representation theory and the established technology acceptance model as an example for an extended study. We conclude by discussing how the example of this research study may inspire IS researchers to overcome traditional theory boundaries and converge rather than diverge existing approaches to IS research in related domains.
Table of Contents
The information systems discipline is relatively new. It evolved at the intersection of historically well-established research fields such as management science, computer science, organisational theory and others (Vessey et al., 2002). Researchers studying in the IS area have mostly originated from one of these reference disciplines, bringing with them not only a range of methods and methodologies but also a diversity of underlying philosophical assumptions about research and, going deeper, regarding understanding and cognition of reality, language and truth. However, since we understand our discipline is concerned with ‘the effective design, delivery, use and impact of information technology in organisations and society’ (Avison and Fitzgerald, 1995), we feel that it is quite uniquely placed at the interface of technology and organisation, unlike some of its foundational or reference disciplines. That is, it addresses the interaction in human-machine systems (Lee, 2001).
The evolution of IS research since its inception has led to the consequence that most of the theories used stem from its reference disciplines. Overall, a wide range of distinctly different foundational theories is being used in IS research, leading to considerable diversity (Robey, 1996; Vessey et al., 2002; Benbasat and Zmud, 2003) and the coining of the term ‘fragmented adhocracy’ (Banville and Landry, 1989) as a description of it. This has also resulted in an ongoing quest for a cumulative tradition, in the hope of evolving a research discipline that builds on an existing body of knowledge, has an awareness of the remaining open challenges, and is guided by a methodological procedure in its future research efforts (Kuhn, 1962; Keen, 1980; Weber, 1997).
The ongoing debate about what constitutes the IS field has centred on the question of what are the core theories unique to information systems that define the discipline and, from a broader perspective, its body of knowledge (Benbasat and Weber, 1996). The argument used in this context is that a reliance on foundational theories from reference disciplines distracts from the main game — namely, identifying, articulating and foremost of all, researching core phenomena that give IS its identity (Weber, 1987). In other words, unless the IS discipline evolves based on a unique core that comprises topics, theories and methodologies, there is a danger of it remaining an intellectual convocation of individuals that pledge allegiance to other disciplines while studying phenomena nominally ascribed to information systems (King, 1993).
Looking at how to address this, Benbasat and Weber (1996) identify three types of diversity, these being diversity in the phenomena that are being studied, diversity in the theoretical foundations that guide such studies, and diversity in the research methods used to study them.
In this paper we reflect on progress in a dedicated subset of the IS field, addressing selected instances of these three types of diversity. We focus on conceptual modelling and its associated phenomena, an area that is widely regarded as inseparable from IS development (Kottemann and Konsynski, 1984; Karimi, 1988). It also has repeatedly been proposed as one of the core artifacts in IS research overall (Weber, 1997; Frank, 1999). In terms of theoretical foundations, we look at the emergence of a promising candidate for conceptual modelling theories, namely models of representation (Wand and Weber, 1990, 1993, 1995) that are referred to as representation theory. In terms of research methods we investigate the principles of representational analysis (Rosemann et al., 2004), also referred to as ontological analysis.
The aim of our paper is to assess the current state of research in this specific domain and to give guidance on how to progress this state. While models of representation and the process of representational analysis per se have been shown to result in interesting and relevant findings, there remains a need for these studies to transcend their current research scope. We argue that representation theory and associated research efforts can be further advanced to exert a wider influence on information systems if this stream of research is put into a broader context by studying the impact of the findings on further phenomena relevant to our research domain. We recapitulate existing scholarly approaches and then discuss a research design that aims at expanding the scope of representation theory and method of representational analysis by converging it with other IS-specific theories, in this instance, the technology acceptance model.
We proceed as follows. In the next section, the main principles of representation theory and previous work in this area are reviewed, with consideration paid to the scope and focus of the analyses. Following this, we present and discuss a research model that converges representation theory and the technology acceptance model and briefly outline the stages of our research. The paper concludes with a discussion of contributions and guidance on how fellow researchers may extend the scope of the theory and related method.