Chapter 1 Information Systems Strategic Management
Chapter 1 examines:
- Why do Information Systems matter?
- How might the domain of information systems be characterised?
- What is the impact of taking a technological view of information systems; how is this changed by a human-centred position?
- From what theoretical perspective is information systems best informed?
- How does information systems appear from a social theoretical perspective?
- The domain of IS may be seen as problematic largely because of the tensions between technology-based and human-centred approaches. Case studies such as London Ambulance and Wessex Area Health highlight some of the problems found in this respect. Whilst methodological attempts have been made at combining approaches from both perspectives it could be argued that, since the domain may be viewed from a perspective of social theory, a social theoretical underpinning such as Burrell & Morgan is needed to give IS the substance it appears to lack.
- The main approach to IS development has been one premised on technology and using substantially the systems development life cycle (SDLC). Alternatives have been pursued from a human-centred and socio-technical perspectives, including for example ETHICS, multiview, and client led design.
- Information Systems Failure
(Lyytinen and Hirschheim, 1987)
Correspondence Failure: The failure of the final ‘system' to correspond with the specification determined in advance.
Process Failure: Failure in the development process, usually in the form of a cost overrun or inability to complete the development.
Interaction Failure: Users fail to use the ‘system' sufficiently, effectively meaning it has failed.
Expectation Failure: Failure of the completed ‘system' to meet the expectations of participants.
Correspondence failure, since it deals only with the correspondence of the finished system to a previously agreed specification, gives rise to a number of problems, of which examples are:
- Considering the other three types of failure it may be seen as an inadequate appreciation of the situation.
- A system which is completed but not used may be seen to serve little purpose.
- What should we do about those systems for which a specification cannot be written
- I would suggest that in order to answer this it would be best to refer to the Burrell & Morgan grid. A hard approach is therefore functionalist in orientation and technology focused; a soft approach is interpretivist and human focused.
- This discussion can be based on the systems development lifecycle, and the extent to which "user requirements" are considered within such an approach. The shortcomings can be difficult to put across to students, since it appears that the SDLC does indeed allow for human interaction within the system. However the key to understanding this is that in essence the systems development lifecycle is a waterfall approached premised on a scientific, instrumental, or functionalist view of the domain of information systems; consequently all human interaction within such an approach ends up being constrained by the functionalist nature. The solution, it is suggested, lies in approaching the domain from a human-centred viewpoint, in much the way that client led design attempts to do.
- It is important to recognise that the Burrell & Morgan typology was specifically designed to explain different social theories, albeit with the target of organisational intervention or analysis. The extent to which such an approach is relevant to information systems is dependent on a view of IS which sees it as informed by human interaction rather than technologically determined. To the extent that Information systems as a domain may be argued to be systems of human activity, the functionalist/interpretivist distinction is therefore of relevance to that domain. The discussion should therefore centre on these issues.
The aim of this exercise is to bring together the learning of the chapter by applying the different approaches to a single case. The organisation concerned is engaged in manufacturing, but seems to be seeing a re-organisation of its human resource as the key to an improved process. Therefore there are clearly human/technical issues to be considered.
- A technology-based approach would, at the extreme, concentrate on the functions and hard tangible issues which form part of the manufacturing process. The basic assumption would be that the manufacturing process could be changed and would function irrespective of the human interaction; in other words, if all the people were changed the process would not suffer. Such an approach could concentrate on, for example, the systems development life cycle method, or on using a wider methodology such as project planning, but in all cases the major consideration would be the technology that was being used. The major benefit of such a method is that, where the system of concern is primarily a technical system, such an approach arguably privileges those elements seen to be of greatest importance. The shortcomings can be brought to the fore by considering the alternative, which would be a system where a technical approach is taken, but in which human activity affects the outcomes.
- The additional benefits derived by a human-centred approach could be assessed by reflecting on the idea of a movement to flexible teams. Clearly the organisation feels that it has the people in-house to enable them to make this change. What a human-centred, participative approach would achieve is the inclusion of all those people within the system of concern so that the developed system could better serve their needs or meet their expectations.
- The key to understanding this problem is that essentially most soft or human-centred approaches are premised on debate, and therefore have outcomes such as agreement or consensus. Whilst this has the effect of enabling a richer understanding of the problem context, it is difficult to see how such methods result in a working system without recourse ultimately to a functionalist, problem-solving approach. There could be discussion of such methodologies as ETHICS, multiview and client led design as a way of approaching this, and ultimately I would look to the tutor bringing to the students attention the idea that what these three methodologies are doing in essence is recognising the range of problem contexts that are represented by the Burrell & Morgan grid, and the need for some means of mixing methods together to match those diverse contexts.
- At this stage the students will have been exposed only to the Burrell & Morgan grid, and so would be expected simply to reflect on the idea that social theory has something to offer, and what the range of social theory is compared to the perspective normally taken on information systems issues.