Introduction
The information systems (IS) field is dominated by positivistic research approaches and theories (Chen & Hirschheim, 2004). IS scholars have pointed out weaknesses in these approaches and theories and in response different strands of post-modern theories and constructivism have gained popularity—see, Lee, Liebenau, and DeGross (1997) and Trauth (2001). The approaches argued for include ethnography, constructivism, grounded theory, and theories like Giddens’ structuration theory and Latour’s actor-network theory. (We refer to these different research approaches and theories as “post-approaches” and “post-theories” when distinction is not required).
background
Although post-approaches and post-theories overcome some of the problems noted with positivistic approaches and theories, they have at least three major weaknesses and limitations. First, their fascination with the voices of those studied leads to IS research as mere reportages and local narratives which can lead to any narrative/reportage being as good as another narrative/reportage. Second, their focus on agency leads to ignoring the structural dimension—the agency/structure dimension is collapsed, leading to a flat treatment of the dimension. Third, their rejection of objectivist elements leads to problems when researching ICT-artifacts and ICT-based IS. For elaborate critique of post-approaches and post-theories, see Archer, Bhaskar, Collier, Lawson, and Norrie (1998).
An alternative to traditional positivistic models of social science as well as an alternative to post-approaches and post-theories is critical realism (CR). CR argues that social reality is not simply composed of agents’ meanings, but that there exist structural factors influencing agents’ lived experiences. CR starts from an ontology which identifies structures and mechanisms through which events and discourses are generated as being fundamental to the constitution of our natural and social reality. This article briefly presents CR and exemplifies how it can be used in IS research.
critical realism in is research
CR has primarily been developed by Roy Bhaskar (1978, 1998) and can be seen as a specific form of realism. Good summaries of CR are available in Sayer (2000) and Archer et al. (1998) and key concepts and main developments are presented in Hartwig (2007). CR’s manifesto is to recognize the reality of the natural order and the events and discourses of the social world. It holds that:
… we will only be able to understand—and so change—the social world if we identify the structures at work that generate those events and discourses . These structures are not spontaneously apparent in the observable pattern of events; they can only be identified through the practical and theoreti -cal work of the social sciences. (Bhaskar, 1989, p. 2)
Bhaskar (1978) outlines what he calls three domains: the real, the actual, and the empirical. The real domain consists of underlying structures and mechanisms, and relations; events and behavior; and experiences. The generative mechanisms, residing in the real domain, exist independently of but capable of producing patterns of events. Relations generate behaviors in the social world. The domain of the actual consists of these events and behaviors. Hence, the actual domain is the domain in which observed events or observed patterns of events occur. The domain of the empirical consists of what we experience; hence, it is the domain of experienced events. Bhaskar argues that:
..real structures exist independently of and are often out of phase with the actual patterns of events. Indeed it is only because of the latter we need to perform experiments and only because of the former that we can make sense of our performances of them. Similarly it can be shown to be a condition of the intelligibility of perception that events occur independently of experiences. And experiences are often (epistemically speaking) ‘out of phase’ with events—e.g. when they are misidentified. It is partly because of this possibility that the scientist needs a scientific education or training. Thus I [Bhaskar] will argue that what I call the domains of the real, the actual and the empirical are distinct. (Bhaskar, 1978, p. 13)
CR also argues that the real world is ontologically stratified and differentiated. The real world consists of a plurality of structures and mechanisms that generate the events that occur.
CR has primarily been occupied with philosophical issues and fairly abstract discussions. In recent years attention has been paid to how to actually carry out research with CR as a philosophical underpinning—see Layder (1998), Robson (2002), Kazi (2003), and Pawson (2006). Gregor (2006) argues that five interrelated types of IS theory can be distinguished: (1) theory for analyzing, (2) theory for explaining, (3) theory for predicting, (4) theory for explaining and predicting, and (5) theory for design and action. The five types can be clustered into two main types: “traditional” natural/social research (first four types) and design science research (fifth type). This section briefly presents how CR can be used in the first four types of IS research and the next section addresses IS design science research based on CR.
Bhaskar says that explanations (theories) are accomplished by the RRRE model of explanation comprising a four-phase process: (1) Resolution of a complex event into its components (causal analysis); (2) Redescription of component causes; (3) Retrodiction to possible (antecedent) causes of components via independently validated normic statements; and (4) Elimination of alternative possible causes of components.” (Bhaskar, 1998). This is a rather abstract description of explanation (theory) development. Here we will instead use Layder’s (1998) less abstract “adaptive theory.” It is an approach for generating theory in conjunction with empirical research. It attempts to combine the use of pre-existing theory and theory generated from empirical data. Figure 1 depicts the different elements of the research process. There is not some necessary or fixed temporal sequence. Layder stresses that theorizing should be a continuous process accompanying the research at all stages. Concerning research design and methods, CR is supportive of: (1) the use of both quantitative and qualitative methods,
(2) the use of extensive and intensive research design, and
(3) the use of fixed and flexible research design.
To exemplify how CR and Layder’s adaptive theory can be used in IS research addressing Gregor’s (2006) four first IS theory types, we will use a project on the use of executive information systems (EIS). The project was done together with Dorothy Leidner.1 Here a new discussion of the research is carried out.
Layder’s adaptive theory approach has eight overall parameters. One parameter says that adaptive theory “uses both inductive and deductive procedures for developing and elaborating theory.” (Layder, 1998). The adaptive theory suggests the use of both forms of theory-generation within the same frame of reference and particularly within the same research project. We, based on previous EIS theories and Huber’s (1990) propositions on the effects of advanced IT on organizational design, intelligence, and decision making, generated a number of hypotheses (a deductive procedure). These were empirically tested. From a CR perspective the purpose of this was to find patterns in the data that would be addressed in the intensive part of the study. We also used an inductive procedure. Although previous theories as well as the results from the extensive part of the project were fed into the intensive part, we primarily used an inductive approach to generate tentative explanations (theories) of EIS development and use from the data. The central mode of inference (explanation) in CR research is retroduction. It enables a researcher, using induction and deduction, to investigate the potential causal mechanisms and the conditions under which certain outcomes will or will not be realised. The inductive and deductive procedures led as to formulate explanations in terms of what mechanisms and contexts could lead (or not lead) to certain outcomes—outcomes being types of EIS use with their specific effects.
Another parameter says that adaptive theory “embraces both objectivism and subjectivism in terms of its ontologi-cal presuppositions” (Layder, 1998). The adaptive theory conceives the social world as including both subjective and objective aspects and mixtures of the two. In our study, one objective aspect was the ICT used in the different EIS and one subjective aspect was perceived effects of EIS use.
Figure 1. Elements of the research process
Two other parameters say that adaptive theory “assumes that the social world is complex, multi-faceted (layered) and densely compacted” and “focuses on the multifarious interconnections between human agency, social activities and social organization (structures and systems)” (Layder, 1998). In our study we focused the ‘interconnections’ between agency and structure. We addressed self (e.g., perceptions of EIS), situated activity (e.g., use of EIS in day-to-day work), setting (e.g. organizational structure and culture), and context (e.g., national culture and economic situation). Based on our data we hypothesized that national culture can affect (generate) how EIS are developed and used and how they are perceived. We also hypothesized that organizational ‘strategy’ and ‘structure’ as well as ‘economic situation’ can affect (generate) how EIS are developed and used and how they are perceived.
Our study and the results (theory) were influenced by, e.g. Huber’s propositions, the ‘theory’ saying that EIS are systems for providing top-managers with critical information, and Quinns’ competing values approach (Quinn et al., 2004). The latter theory was brought in to theorize around the data from the intensive (inductive) part of the study. Adaptive theorizing was ever present in the research process. In line with CR, we tried to go beneath the empirical to explain why we found what we found through hypothesizing the mechanisms that shape the actual and the events. Our study led to our argument that it is a misconception to think of EIS as systems that just provide top-managers with information. EIS are systems that support managerial cognition and behavior—providing information is only one of several means—as well as it can be one important means in organizational change. Based on our study, we “hypothesize” that “tentative” mechanisms are, for example, national culture, economic development, and organizational strategy and culture. We also hypothesized how the mechanisms together with different actors’ decisions and actions, based on their desires, beliefs, and opportunities, lead to the development and use of different types of EIS. For example: (1) EIS use for personal productivity enhancement respectively EIS use for organizational change, and (2) EIS use for organizational change respectively EIS use for control and stability.
future trends
This section presents how CR can be used in IS design science research. The primary constituent community for the output of IS design science research is IS-professionals (Walls et al., 1992). This means primarily professionals who plan, manage and govern, design, build, implement, operate, maintain and evaluate different types of IS initiative and IS.
Using van Aken’s (2004) classification we can distinguish three different types of designs an IS professional makes when designing and implementing an IS-initiative: 1) an object-design, which is the design of the IS intervention (initiative), 2) a realization-design, which is the plan for the implementation of the IS intervention (initiative), and 3) a process-design, which is the professional’s own plan for the problem solving cycle and includes the methods and techniques to be used to design the solution (the IS intervention) to the problem. IS design science research should produce knowledge that can be used by the professionals in the three types of designs. Van Aken defines a technological rule as “.an instruction to perform a finite number of acts in a given order and with a given aim”; and a technological rule is “a chunk of general knowledge, linking an intervention or artefact with a desired outcome or performance in a certain field of application” (van Aken, 2004, p. 228). A technological rule is general, which for IS design knowledge means that a rule is a general prescription for a class of IS problems. Since a technological rule should be used by practitioners it should be applicable and actionable. Generally, the form of the technological rules is like “if you want to achieve A (outcome) in situation B (problem) and context C, then something like action/intervention D can help because E (reason).” “Something like action/intervention D” means that the rule is to be used as a design exemplar. A field-tested and grounded technological rule has been tested empirically and is grounded in science. Field-tested and grounded technological rules will in most cases be in the form of heuristics. This is consistent with CR’s view on causality and means that the indeterminate nature of a heuristic technological rule makes it impossible to prove its effects conclusively, but it can be tested in context, which in turn can lead to sufficient supporting evidence (Groff, 2004).
Van Aken (2004) suggests that management design science research has much in common with CR-based evaluation research of social programs (Pawson & Tilley, 1997; Kazi, 2003). In line with CR-based evaluation research, the intention of IS design science research is to produce ever more detailed answers to the question of why and how an IS initiative works, for whom, and in what circumstances. This means that a researcher attends to how and why an IS initiative has the potential to cause the (desired) change. In this perspective, an IS design science (ISDS) researcher works as an experimental scientist, but not according to the logics of the traditional experimental evaluation research. Bhaskar states: “The experimental scientist must perform two essential functions in an experiment. First, he must trigger the mechanism under study to ensure that it is active; and secondly, he must prevent any interference with the operation of the mechanism. These activities could be designated as ‘experimental production’ and ‘experimental control’.” (Bhaskar, 1998). Figure 2 depicts the realist experiment.
ISDS researchers do not perceive that IS initiatives “work.” It is the actions of different stakeholders and participants that make them work, and the causal potential of an IS initiative takes the form of providing the reasons and resources to enable different stakeholders and participants to “make” changes. This means that an ISDS researcher seeks to understand why and how an IS initiative, for example, the implementation of an enterprise system works through understanding the action mechanisms. It also means that an ISDS researcher seeks to understand for whom and in what circumstances (contexts) an IS initiative works through the study of contextual conditioning.
Figure 2. The realist experiment
ISDS researchers orient their thinking to context, mechanism, outcome pattern configurations (CMOCs). This leads to the development of transferable and cumulative lessons from ISDS research. A CMOC is a proposition stating what it is about an IS-initiative which works for whom in what circumstances. A refined CMOC is the finding of an evaluation of an IS initiative. Outcome patterns are examined from a “theory-testing” perspective. This means that an ISDS researcher tries to understand what the outcomes of an IS initiative are and how the outcomes are produced. Hence, the researcher does not just inspect outcomes in order to see if an IS initiative works, but analyzes the outcomes to discover if the conjectured mechanism/context theories are confirmed.
In terms of generalization, an ISDS researcher through a process of CMOC abstraction creates “middle-range” theories. These theories provide analytical frameworks for interpreting differences and similarities between classes and sub-classes of IS-initiatives. Given that the goal is to develop design theories and knowledge—to construct and test CMOCs explanations—for practitioners ISDS researchers need to engage in a learning relationship with IS practitioners.
ISDS research based on the above can be carried out through an IS design science research cycle (Figure 3).
The starting point is theory and problems or issues. The research is driven by problems or issues. Problems or symptoms can be identified by practitioners or by researchers. For example, an organization can have the problem that their “ERP-projects are not leading to desired outcomes.” The problems can also be identified through quantitative studies carried out by a researcher. For example, the researcher can analyze a data base containing use data for an IS and is looking for unwanted patterns. The theory includes propositions on how the mechanisms introduced by an IS intervention into a pre-existing context can generate (desired) outcomes. This entails theoretical analysis of mechanisms, contexts, and expected outcomes. This is the first step in developing technological rules and means that one tries to generate technological rules using our current knowledge, that is, grounding in theory. In general, the IS-researchers have been far from good in systematic reviews of research results. Pawson (2006) shows, from a critical realist perspective, how to do systematic reviews and make sense of a heterogeneous body of literature. Using Pawson’s approach it should be possible to test and refine IS interventions. For example, it is possible to move away from the many one-off studies in the IS-field and instead learn from fields like medicine and policy studies on how to develop evidence-based IS design knowledge. Such a systematic review can be part of the starting point.
The second step consists of generating more specific “hypotheses.” Typically the following questions would be addressed in the hypotheses: 1) what changes or outcomes will be brought about by an IS intervention (initiative), 2) what contexts impinge on this, and 3) what mechanisms (social, cultural, and others) would enable these changes, and which one may disable the intervention. In this step the technological rules are refined.
The third step is the empirical test. It is done through intervention and guided by theory and technological rules.
Figure 3. The information systems design science research cycle
The step includes also the selection of appropriate data collection methods. ISDS research employs no standard research design formula. The base strategy is to develop a clear theory of IS initiative mechanisms, contexts and outcomes. Given the base strategy, an ISDS researcher has to design appropriate empirical methods, measures, and comparisons. In this step it might be possible to generate support of the IS intervention’s ability to “change” reality. Based on the result from the third step, we may return to the IS intervention to make it more specific as an intervention of practice. Next, but not finally, we return to theory. The theory may be developed, the hypotheses and the technological rules refined, the data collection methods enhanced, etc. To develop the technological rules means that the cycle will be repeated. As said above most of the technological rules will be heuristic. Through multiple case-studies one can accumulate supporting evidence which can continue until “theoretical saturation” has been obtained. The researcher can be more or less active in the implementation (use) of the technological rules. The researcher can be very active and work like an action researcher, but can also be quite passive and work like an observer.
conclusion
Although CR has influenced a number of social science fields, it is almost invisible in the IS field. CR’s potential for IS research has been argued by, for example, Carlsson (2003, 2004, 2006), Dobson (2001), Mingers (2004), Mutch (2002), and Longshore Smith (2006). This article argued that CR can be used in IS research—behavioral and design science—to overcome problems associated with positivism, constructivism, and postmodernism.
key terms
Constructivism (or Social Constructivism): Asserts that (social) actors socially construct reality.
Context-Mechanism-Outcome Pattern: Realist evaluation researchers orient their thinking to context-mechanism-outcome (CMO) pattern configurations. A CMO configuration is a proposition stating what it is about an IS initiative which works for whom in what circumstances.A refined CMO configuration is the finding of IS evaluation research.
Critical Realism: Asserts that the study of the social world should be concerned with the identification of the structures and mechanisms through which events and discourses are generated.
Empiricism: Asserts that only knowledge gained through experience and senses is acceptable in studies of reality.
Positivism: Asserts that reality is the sum of sense impression. In large, equating social sciences with natural sciences. Primarily using deductive logic and quantitative research methods.
Postmodernism: A position critical of realism and rejects the view of social sciences as a search for over-arching explanations of the social world. Has a preference for qualitative methods.
Realism: Aposition acknowledging a reality independent of actors’ (incl. researchers’) thoughts and beliefs.
Realist IS Evaluation: Evaluation (research) based on critical realism aiming at producing ever more detailed answers to the question of why an IS initiative works (better) for whom and in what circumstances (contexts).
Retroduction: The central mode of inference (explanation) in critical realism research. Enables a researcher to investigate the potential causal mechanisms and the conditions under which certain outcomes will or will not be realised.
