A model system or controller can only model or control something to the extent that it has sufficient internal variety to represent it. (Law of Requisite Variety or Ashby’s Law[1])
Everything should be made as simple as possible, but not simpler. (Albert Einstein)
In support of the view that exploratory practice-driven research is high potential and under-utilised, this paper reports on a seven-year research program to improve IS project management. The researchers have conducted a series of engagements with practitioners to improve practice through the development of new theory. Each of the research studies draws on a different theoretical framework and generates different insights while building on the earlier engagements to improve project performance. The different frameworks provide the necessary variety, in Ashby’s terms, to model and control improvements in project performance.
Examining the IS project management challenge early in the research program, the researchers concluded that there were at least three explanations for the failure to develop effective theories through which to manage large and complex IS project performance:
Either the research community has failed to identify the right factors or processes, or factor and process models are insufficient by themselves;
Project performance is subject to high contextual complexity and multiple contingencies; and
Researchers have looked in the wrong places, wearing the wrong glasses.
All three explanations suggest the challenge needs reframing, reinforcing the search for new theory.
|
Objective |
Step |
Description |
References |
|---|---|---|---|
|
Engagement |
Form a multi-disciplinary team |
Access to multiple frameworks |
Pettigrew (1990) |
|
Immerse researchers in practice |
Role in practice, direct observation, participant observation |
||
|
Looking for the gap in practice[a] |
Choose the ‘gap in practice’ |
What is the unusual behaviour? A prepared mind ‘The problem of the problem’ |
Benbasat and Zmud (1999) |
|
Articulate the problem |
What is done:
|
Klein and Myers (1999) |
|
|
Finding the theory in that gap |
Understand the ontological and epistemological underpinnings |
Consider the strengths and limitations of the meta-theoretical assumptions that have either explicitly or implicitly been adopted What are the strengths and limitations of their implicit assumptions? What embedded assumptions in the world might be relaxed (testing assumptions)? |
Weber (2003) |
|
Form the evoked set of theories |
Look at data in the gap to signal which things will be useful to point us towards the theories Examine perspectives from other fields
|
Whetten (1989) |
|
|
Develop a combination and/or permutation of theories |
Which theories shed light on the gap or are dispensable? Can multiple theories integrate? |
||
|
Account |
Develop an account of the phenomena |
The explanation of the hypothesised laws:
|
Weber (2003) |
|
[a] The choice and articulation of the phenomena often occur concurrently rather than as discrete sequenced event. Weber, R. 2003, 'Editor's comments: Theoretically speaking', MIS Quarterly, vol. 27, no. 3, pp. iii-xii. |
|||
Table 3 presents the research approach. The approach is iterative within each step and adopts Burrell and Morgan’s (1979) functionalist paradigm, which views social science as objective and ordered, reflecting the researchers’ positivistic orientation.
Practice-driven research involves a collaborative effort between a research team and the sponsors of the research effort (Zmud, 1998). This research program consists of a series of engagements with organisations, from both the public and private sectors, undertaking a program of IS-based business change. Engagements extended over a period of three to five years and involved deep immersion of the research team in the organisations followed by periods of reflection and theory development.
Engagements were conducted within various research frameworks, including single-case, multi-case, and longitudinal case studies (Yin, 2003); grounded theory (Eisenhardt, 1989; Strauss and Corbin, 1997) and action research (Baskerville and Pries-Heje, 1999; Susman et and Evered, 1978) .
The engagements provided theory-driven frameworks and recommendations. Researcher participation has ranged from direct observation in executive steering committees to participant observer (Jorgenson, 1989; McCall and Simons, 1969) undertaking project roles as well as being part of the research team. The extent of participant observation provided a unique perspective of operations across the organisations and extensive access to the research subjects. Understanding increases by being there as part of the project control system.
Formal data collection protocols applied to three primary forms of data. First, semi-structured interviews with individual informants were recorded, transcribed and validated. Second, direct observation augmented, compared and corroborated evidence in meetings, reviews and informal gatherings. Third, documents provided information on data gathered from interviews. These documents included strategic plans and business plans, proposals, reviews, policy and procedure manuals, release plans, project plans and specifications, reports, letters, minutes, memoranda and media clippings. Together, these multiple sources of data enabled triangulation of evidence (Carson et al., 2001).
The research background included organisational psychology, philosophy, political science, marketing, systems design and engineering. Some of the researchers have held senior positions in industry. This diverse set of theoretical and practical backgrounds enabled open dialogue and simultaneous engagement in robust debate with senior managers and between the researchers.
Weber (2003) describes the choice and articulation of the phenomena to be explained or predicted via theory as the two most critical tasks undertaken by researchers.
When undertaking exploratory practice-driven research there are two characteristics that help to identify a gap in practice. The first is the absence of practice predicted by current theory. The second is observing practice that is inconsistent with current theory. Together, they strongly suggest that current theory is inappropriate as a basis from which to resolve the problem. Studying such events, which, for some reason, have behaved differently from what established knowledge would prescribe, is consistent with support for in-depth research in a single organisation (Sauer et al., 1997).
The identification of practice-driven research problems requires a prepared mind. Otherwise, the researcher simply treats departures from expectations as errors whereas the research team must be sensitive to such departures and assess them against their different theoretical backgrounds to identify unexpected insights. To do this, problem statements must be clearly articulated. Weick (1989) highlights that:
… the problem statements that drive the theorising process are more complex than they appear to be. Not only do they contain an anomaly to be explained, but they also contain a set of assumptions that can be confirmed or disconfirmed.
They require a description of what is problematic, the specific characteristics of and assumptions about the context, and identifications of limits to the domain.
A final challenge in practice-driven research is that the sponsors must also agree on the problem, with the sponsors often subject to stringent time requirements.
There is an extensive literature on what constitutes good theory but limited guidance on good theorising or how to develop good theory. Developing new theory to account for practice ‘commonly involves borrowing a perspective from other fields, which encourages altering our metaphors and gestalts in ways that challenge the underlying rationales supporting accepted theories’ (Whetten, 1989).
A useful place to start is to develop an understanding of the ontological and epistemological underpinnings and look at what researchers have taken for granted. It is then possible to challenge or relax the most accepted propositions in the current theory and to explore alternative explanations of the phenomena. To do the latter (that is, fit an alternative theory to the problem), researchers look for a theory, or theories, that simultaneously define the gap and account for the features in the gap.
Good guidelines are available for developing a theoretical account of a phenomenon. See, for example, Weber (2003) and Weick (1989). Weber (2003) describes this step as the explanation of the laws that are hypothesised, including their constructs, interactions, states and lawful transitions.