George Bakehouse

Empirical Research in Information Systems

The first section of this paper provides a view of the relationship between research and reality. Central to this debate is the notion of two separate and sometimes competing approaches to solve real life problems, theory and practise. The last three or so decades has witnessed the development of numerous methodologies which vary across a wide spectrum from the “very hard” to the “very soft” most claiming to have practical benefits in the real world. An area of Systems Science that has grown quickly amongst all the confusion is that of Information Systems, a new and highly dynamic subject area where academics and practitioners often fail to agree at any level about things as fundamental as the meanings of ‘information’ and ‘system’. There is ample evidence to show that in the world of Business Information Systems, technologists do not understand the world of business and vice-versa. The final section of this paper describes an ongoing action research program that spans several sectors of commerce and industry. The methods, tools and techniques adopted in this study are drawn from numerous paradigms and disciplines; many have been specifically adapted for the research program. These tools and techniques range from participant observation, SWOT analysis, information problem classification, phenotypes of erroneous action, human error modelling through to psychometric testing accompanied by “standard” information systems tools. Central to the research program is the concept of building links between research and reality via empirical research.

Research and Reality: Co-Exist or Co-Inhabit?

Our ongoing empirical research compares leading technological organisations in four sectors of the UK economy; these are Banking (Citicorp), Construction (Kvaerner — Trafalgar House), Health (Frenchay NHS Healthcare Trust) and Transportation (LEX). This comparison identifies their stages of IS development, their relative timescales and costs (measured in terms of IS investment per employee per annum) and their information quality (indicated by the average number of defects suffered by each employee each day). A goal is to improve information quality control by back-tracking the causes of defects and evaluating their effects by forward-tracking, where possible.

To Err is Human

The first section of this paper will summarize an ongoing action research project currently being undertaken by the author and a team of academics based at the University of the West of England (UWE, Bristol) spanning nearly a decade. This research project has developed an approach to embedding the tools and techniques of systems engineering in an action research framework (Bakehouse et al., 1995, 1997; Doyle, 1994; Waters et al., 1994). These projects have involved strategic, tactical and operational systems in transport (Lex Transfleet), health care (Frenchay Health Trust, Neurosciences) construction (Trafalgar House, WPE Homes), banking (Citibank), and other areas of the private sector. The theories, tools, techniques and methods adopted for the research program were selected on the basis of their relevance to the solution of real problems discovered in everyday working environments. The initial phases established a methodology that identifies, quantifies, and classifies information problems. The author progressed the research by developing a taxonomy of phenotypes of erroneous action that classified the “causes” of the information problems in terms of Human error.

Research and Reality

This paper will begin with a discussion of the relationship between research and reality. Central to this debate is the notion of two separate and sometimes competing approaches to solve real life problems. This debate is of particular significance to the information systems discipline which is in its relative infancy. For the last few decades information systems development methodologies have been numerous and varied ranging from the “Hard” to the “Soft” approach. This paper IS NOT yet another attempt to compare and contrast extreme examples from the continuum of methodologies, rather a gentler approach is taken to explore the underlying problems of relating research and reality irrespective of ideological “camps”.

“Man Is a Creature Made at the end of the Week… When God Was Tired”: Some Reflections on the Impact of Human Error Upon Information Systems

The notion of Human error has different meanings for many disciplines. Cognitive theorists see them as an important clues to the covert processes underlying routine human action. To applied practitioners they are the main threat to the safe operation of high risk systems. The theoreticians like to collect, cultivate and categorise errors, practitioners are more interested in their elimination and where total elimination is not possible in containing their adverse effects as much as possible. This paper will concentrate on the theoretical and applied practical approaches putting theory into practice. Identification and classification of Human error is essential in the design of information systems whether computer based or manual if the elimination of error is to be achieved. As Mach (1905) so aptly stated “ Knowledge and error flow from the same mental sources, only success can tell the one from the other.”

Putting Systems Theory into Practice and Vice Versa

Excerpt only. For full access, check out the book through your local library, request it on interlibrary loan, or order it through a book dealer.