chapter5

	E I P	5: IT Management Structures Christine Page-Hanify and Kim Johnston
		[next chapter] [previous chapter] [contents]

This chapter describes and analyses the structures which support the development and management of information technology within universities. It finds that the IT management structures which currently exist are based on different organisational models each with their own particular strengths and weaknesses. It further identifies some common trends in structural change across the sample, and finds that technology convergence is influencing the shape of these structures. In addition, because the application of IT is becoming increasingly important, both in terms of core operations of the university as well as its proportion of the budget, the chapter recommends that the IT general manager or 'chief information officer' of the university play an executive role in contributing to the development of university policy, planning and decision making.

Approaches to the organisation and management of IT activities have typically been characterised on a continuum from a centralised model, where the centre has dominant influence over both the current IT activities and future direction to one in which divisional and business unit management has most responsibility for IT, to complete decentralisation of control both operationally and strategically. Developments in technology itself have facilitated the changing mode of structural alignment of IT activities, with systems development, operations and support being increasingly devolved and managed by the end user of the system. This is in stark contrast to the 'glasshouse' of the mainframe era, when all activities were centralised. Many large companies today have attempted to implement federal IT management structures - a division of powers with responsibility for IT distributed to business units combined with some central functional leadership regarding architectures, standards and large IT investments.

Current Environment

The universities visited in our study have an overall similarity of structure with variations of portfolio responsibility for technology and staff development which influence the effectiveness of service delivery. They each have some governing board, a Chief Executive Officer (Vice-Chancellor), an academic stream with a minimum of one Deputy Vice-Chancellor and an administration stream with a minimum of a Registrar. The size and diversity of the institutions affect the number of Deputy Vice-Chancellors, Pro Vice-Chancellors, and senior administrators. All had a complex array of special purpose academic committees, several related to technology covering operational, policy and strategic issues. The location of Library, IT Services, Educational Technology and Staff Development report generally on traditional lines (i.e. Library and Educational Technologies to Academic management, IT Services to Administration) while Staff Development is typically split into two streams, academic and non-academic groups. We discuss the increasing convergence of these activities later in this chapter.

Within that common framework Australian universities fall into two broad types of organisational structures, which have tended to reflect their origins and history. The older universities have been shaped by a professional bureaucratic model in which considerable operating autonomy has typically been devolved to faculties, departments and individual academics, and examples of this structure include the Universities of Melbourne, Queensland and New South Wales. The newer universities, such as South Australia, Griffith and Wollongong, are characterised by more centralised bureaucratic structures. These different structures influence the way in which the institutions react to environmental influences and pressures for change. Importantly, they also influence the form of IT management adopted. While both organisational forms have faculties operating as quasi-divisions, the key difference is the extent of devolution of decision making and budgets to those divisions. In the older established universities, the delineation between administrative and academic boundaries is more significant and this is reflected in industrial arrangements and the provision of support services. These differences between the types of university structures are reflected in how they manage IT. Before proceeding to analyse the strengths and weaknesses of these different structures, we describe the evolution of IT structures generally, and then describe existing IT management structures in some of the universities studied.

Technology Change and the Evolution of IT Structures

In many organisations, information technology units have evolved in response to the changing role of technology in the workplace. The 'first generation' in the late 60's and early 70's was a wholly contained group of technical professionals who managed all elements of the computer operation. The applications were principally supporting the automation of accounting and other process tasks such as student administration. The structure was therefore very simple and the interface to the organisation limited. The second generation saw the establishment of specialist technical groups within the IT unit as it grappled with the expansion of tools, methods and systems (e.g. a specialist database unit, networks group), and a more complex link into the organisation as the applications increasingly supported the delivery of services to customers. The current change results from the IT professional needing to work more closely with the expert or knowledge worker in applying technology to the core business of a profession. This is the stage (or transformation process) that is underway in universities through the introduction of educational technology.

There is a lot of bottom up activity into looking at how [the academic's] teaching can be improved through technology.

Now, IT units are forming flexible multi-disciplinary teams to develop and support the new generation of applications, while centralising 'routine' activities (further supported by automation). Thus structures have evolved to some extent in response to developments in technology, from a central 'mainframe shop' to a distributed, client-focused work-force, where the skilled IT staff are dispersed to the areas where the technology peripherals exist and the centre, if at all, retains a small technical support group that coordinates the support activities. This is the model required to support the development of educational technology.

University Technology Management Structures

To provide some examples of where universities locate IT management in their organisation structure, we briefly describe the top-level IT management structures at three universities below, as at November 1996.

Case Study 7: The Australian National University
Reporting to the Vice-Chancellor of The Australian National University are:

Deputy Vice-Chancellor(DVC)

Director of the Institute of Advanced Studies and DVC

Pro Vice-Chancellor(PVC) and Chair of the Board of the Faculties

Pro Vice-Chancellor and Chair of the Board of the Institute of Advanced Studies

Pro Vice-Chancellor(Planning and Administration)

Pro Vice-Chancellor(Finance and Development)

All IT is managed by the IT Director who reports to the Director, Institute of Advanced Studies and Deputy Vice-Chancellor. Each of the six units (Management Information Services Division, Teaching & Learning Technology Support Unit, ANU Supercomputing Facility, Centre for Networked Information and Publishing, IT User Support, Network Services) have their priorities and project funding coordinated through an advisory committee process. Planning and Policy for the Division operates outside the committee structure. Each unit has its own committee and includes the Director and the Unit Manager. The Presiding Members of each committee form the Information Technology Strategy Committee which sends recommendations on the budgets and priorities for the IT Units to the Australian National University's Budget Advisory Group. The IT budget is comprised of two elements, a central component and a percentage off the top for which they bid through the committee process.

The Library also has its own budget and the Librarian reports to the Deputy Vice-Chancellor. Coordination between the Director IT and the University Librarian is encouraged but the development of combined strategies is often difficult to implement.

Provision of technology in teaching spaces is the responsibility of the central administration.

Academic staff development is provided through the Centre for Educational Development and Academic Method (CEDAM) who have been encouraging the introduction of educational technologies through seed funding to support innovation. Participants are required to share their results through workshops. Recently an academic from Computing Science has been seconded to the Centre to support the educational technology initiatives. Other forms of staff development are at the discretion of the budget unit. At the time of the site visit CEDAM reported to the Registrar; however, the Australian National University is currently transferring responsibility to the Pro Vice-Chancellor and Chair, Board of the Faculties. Within that portfolio, the three units, Dean of Students, Centre for Continuing Education and CEDAM are expected to become a combined division.

Case Study 8: Queensland University of Technology
Reporting to the Vice-Chancellor of Queensland University of Technology are:

Deputy Vice-Chancellor;

Pro Vice-Chancellor (Academic);

Pro Vice-Chancellor (Research and Advancement);

Executive Office; and

Budget and Planning.

The Pro Vice-Chancellor (Information Services) reports to the Deputy Vice-Chancellor and has portfolio responsibilities for Library and Computing Services, including:

communications, telephony, training, PC and laboratory support, administrative systems and web development;

Audiovisual Services;

Computer-based Education;

Education Television; and

Open Learning Unit.

A recent quinquennial review has recommended the merging of the latter four educational support units into a larger one of Teaching and Learning Support Services.

Academic and general staff training are provided by the Academic Staff Development Unit and a General Staff Development Unit (GSDU) respectively and their activities are coordinated by the Senior Management Development Program. The training section works collaboratively with the IT Unit. The GSDU reports to the Registrar although this is under review and may move into the portfolio of the Pro Vice-Chancellor (Information Sciences) as a way of rationalising all training.

Policy is developed through the IT Advisory Committee which is chaired by Pro Vice-Chancellor ((Information Sciences) and supported from CSD. It has representatives of all faculties and Registrars. Technology is also discussed at the Academic Board, Planning and Resources Committee, Vice-Chancellor's Advisory Committee, and the Executive.

The Library also provides training in advanced information retrieval as part of their Ayers Program to train postgraduates in information retrieval which is woven into course-work and shaped to students needs, and Internet training for academics and general access to CD-Rom laboratories.

Case Study 9: The University of Queensland
Reporting to the Vice-Chancellor of The University of Queensland are:

Deputy Vice-Chancellor; and

Secretary and Registrar.

Reporting to the Deputy Vice-Chancellor:

six Pro Vice-Chancellors for the six academic groups; and

Pro Vice-Chancellor Academic Services.

Reporting to the Committee of Pro Vice-Chancellors are:

Research and Postgraduate Studies; and

External Affairs.

Reporting to the Secretary and Registrar:

Central Administration.

The Pro Vice-Chancellor (Academic Services) who provides advice directly to the Vice-Chancellor, reports to the Deputy Vice-Chancellor and has a portfolio that includes the Library, The Prentice Centre, Educational Multimedia Services (EMS), and the Tertiary Education Development Institute (TEDI). The portfolio also covers other non-technical responsibilities.

The Prentice Centre has some core funding but generates significant funds for infrastructure development through fee for service policies. It includes a production group (TV, multimedia and Web application and development); an operations group for facilities management; an infrastructure group supporting engineering, technical and networks; audio-visual for program lectures; an internal financial/admin group from the Centre; and the computer shop. At the time of the site visit this particular configuration was being reviewed and a two group structure was being explored which would support all client services into one and all infrastructure and technology expertise in the other. The Prentice Centre also does its own training.

The Library offers training in information skills, including postgraduate students. They also have their own computer training rooms. They recently developed jointly with Computer Science the Queensland Information Kit (QIK) which is an Internet based self instruction tool.

EMS offers training to administrative staff on a fee for service basis. TEDI runs new staff courses, the Vice-Chancellor's induction course, how to use e-mail and telephones etc. The IT Management Plan has recommended that training be coordinated by Academic Services.

Trends Observed

The above descriptions are primarily illustrative, with the IT management structures of nine other universities also being studied. Three significant trends were evident from analysis of these IT management structures. First, and reflecting the increasing importance of IT in university budgets and operations, a few universities (e.g. The University of Melbourne, the University of New South Wales) are moving responsibility for IT into the senior executive group, with a Deputy Vice-Chancellor or Pro Vice-Chancellor responsible for IT reporting directly to the Vice-Chancellor. Second, in response to the digital revolution and the increasing convergence of computing, communications and information services, a number of universities (e.g. University of Canberra, Queensland University of Technology, Griffith University, University of New South Wales, University of Queensland) are beginning to combine responsibility for IT services, library, multimedia production and, in some cases, staff development. Third, and less evident from organisation charts and top-level structures, there are differences in the extent to which IT services are centralised or devolved. In some universities, such as University of South Australia and Deakin University, there is a strong emphasis on a centralised IT infrastructure. In others, such as the University of New South Wales and The University of Melbourne, there is considerable devolution of IT services to the faculties and schools. We go on to discuss these different IT management models, and how new models are emerging, later in the chapter.

Shifting Responsibilities in University IT Organisations

Within many universities, the systems development group at the centre principally supports the core administrative systems (Financial, Student Systems, Human Resource and Facilities Management). They still operate in 'mainframe' mode. However, systems development of faculty based needs, either administrative or for teaching are often done at the faculty level. Thus the faculties often have significant autonomy over IT development and support for their own specific user needs. This is more evident in the 'group of eight' universities such as The University of Melbourne, The Australian National University and The University of New South Wales.

At UNSW there are many faculty based systems that take data from the Student Information System and feed into systems that support tutorial and class management.

Here at the AGSM [Australian Graduate School of Management], systems development and support to meet our particular needs are undertaken quite independently from the UNSW centre.

Most institutions have IT support split into two - the administrative systems and the academic systems. The administrative IT units are responsible for delivering infrastructure such as networks and telephony (although some still have telephony separate from IT). The trend to merge data and voice networks emerged during the 1980s as technology supporting those services converged. In the early 1990s a further extension of this process included image where video/TV services, supported by digitisation of their medium, converged with other communication structures. Therefore, those universities that have a TV production or a distance education technology capability have considered structural co-location with either academic staff development and/or communications units. A Manager of Education Television said:

We originally provided more production support for journalism in the School of Media, but [we] shifted and became part of the Division of Information Services, as we had a secondary role to support the rest of the community.

Some have also reviewed the location of electronic publishing (the Australian National University and Griffith). Academic computing support units include PC installation and maintenance, computer laboratory support, high performance computing, training, and web design and management. Educational technology services such as audio-visual and television production, multimedia production, desktop publishing, graphic design, and some aspects of instructional design are found either under academic computing support or in academic development units.

Restructuring

One significant structural trend occurring within universities is the amalgamation of Library and IT areas and, in some universities, the inclusion of educational technology areas. Convergence was initially mostly about libraries re-aligning with computing services, as proposed by the Follett Report (1993), because they were both focused on information services. In the past, this convergent activity was focused on electronic information sources and services. There is an increasing awareness of the implications of convergence for courseware development and the process of teaching and learning. While often focused on reducing costs, these moves have also provided an opportunity for restructuring to improve the effectiveness of coordination across related areas. The struggle occurring at present is about who owns educational technology. Is it the Academic Pro Vice-Chancellor or the IT Pro Vice-Chancellor? Each institution is resolving this problem differently. Whether a librarian, an academic or a technologist controls the convergent structure may depend on many factors, but can lead to quite different managerial perspectives.

One third of Australian universities have implemented a convergent model to varying degrees. Very few have integrated the administrative and academic units of IT support, and continue to retain sub-unit structures along client boundaries. It is difficult to introduce multi-disciplinary teams of mixed IT professionals and academics in universities due to the rigid nature of job descriptions, classifications systems and the historic gap between administrative and academic endeavour. This rigidity is being gradually removed from other public sector organisations, in order to adequately deal with both the complexity of, and the changing skill requirements for, IT systems development.

It is also worth noting that the level of convergence in the United States is far less than that in the United Kingdom and Australia, because the United States has not experienced the same level of scarcity of resources, which here and in the United Kingdom has driven structural change. In fact, most convergent structures have been the result of performance reviews of one element or the other, usually with complaints of unsatisfactory performance by faculty, and improved management control and direction was the primary driver. While the faculty and professional structures remain strong relative to the executive, institutional change is not easily effected by strategic means. In Australia, the amalgamation of institutions in 1990 facilitated significant structural change. In the United Kingdom there were clear financial, demographic and market forces as the major change agents. Indeed, in both Australia and the United Kingdom, convergence has been associated with the newer universities. However, a number of the older and more traditional universities are implementing convergent structures (e.g. Birmingham, Edinburgh, Queensland and New South Wales), which may indicate that this is a significant trend.

The IT Support Crisis

As IT is enabling many of the changes occurring in higher education, the load on IT services and support is increasing. At the same time, budgets are decreasing while demand for quality and responsiveness is increasing. Winship (1996) highlights this support crisis:

Teaching staff need support services such as media specialists, instructional systems developers, computing consultants and equipment technicians to assist them in gaining access to and effectively using IT. A phenomenon confronting many universities, however, is an insatiable demand for information technology services which appears to be growing day by day. It includes demands for more dial-in lines, the exponential increase in use of the Internet and the World Wide Web and requests for more and more central and departmental support for desktop computers.

The result is that many universities report that there is a 'siege' environment in their IT support organisations; in that enormous pressures are being generated on the support staff involved on the one hand and that there is a lack of understanding by end users of the complexities involved in their simple request 'to fix it' on the other. Many universities report an 'IT support crisis' brought about by:

a new dependency of academic and administrative work on IT. Academics and general staff rightly perceive that the use of IT is central to their mission and want IT services to deliver on their expectations;

new and expanding sets of customers, often including ones with little or no interest in technology per se;

new customers who are not 'expert' but want authority for decision-making and refuse to be advised by some lowly paid 'techo';

increased per capita demand for services in all aspects of technology support;

a mismatch between expectations and funding at all levels;

a failure or inability of end users and units to assume responsibility for many support functions;

the recognition that front-line support provided by a central IT support unit is not as effective as more locally based support; and

the complexity of support for multi-vendor, distributed IT and the failure to recognise that individuals cannot be experts in everything related to IT.

To meet these compounding demands, the IT organisation continues to change through restructuring to improve the relationships with other business line managers, making the chief information officer (CIO) a formal or informal member of the top management team, out-sourcing various functions, and/or the adoption of an overall technology architecture.

The above influences on the organisation and its IT services unit reveal themselves as a series of dilemmas or significant issues that are being considered and managed in different ways. While administrative systems and network infrastructure are high budget items, there are already mechanisms in place to deal with them. They fit into mainstream IT management, and existing IT structures could deal with these without significant intervention. However, all universities visited stated that the application of technology to teaching and learning has the highest priority.

It is seen as a highly complex problem which affects teaching quality and also requires extensive IT investment. There is concern that they must be more strategic in managing the investment in this element of technology, not least because it is viewed as a means to enter new markets. For example, at The University of Queensland:

The level of activity [in educational technology] is at a point where the Teaching and Learning Committee has to take an active role but not to control. It needs to manage and get a better investment. The IS strategy is the first sign of the corporate view with an eighteen month to two year implementation.

This issue crosses the traditional segmented boundaries of pedagogy and administration. Responsibility for support is also fragmented as sources of funding varies (i.e. research infrastructure, operating funds and other sources). This is due to the autonomous role that a researcher has over projects and funding and the relative autonomy of the faculties from the corporate centre of the university.

It is also clear that the nature of information technologies and the effect they are having on teaching and learning practices is posing particular challenges to the arrangements for supporting staff and students using the technologies. Louise Moran of the University of South Australia:

It changes how you think of centralisation versus decentralisation. The tradition of having a central resource infrastructure group is OK, but a central user services group is no longer useful, except that you need effective ways of allowing local people to get the services they need. It is difficult to grasp and to make happen. The relationship between the Library and ITU and Staff Development has to become closer. It is all very interrelated like a honeycomb.

The Head of the Multimedia Education Unit (MEU) at The University of Melbourne sees the location of educational technology support in an academic unit as an important issue:

I don't believe the average academic has the time to do the scripting. Only the aficionados will do the scripting and they can sometimes trigger a revolution in their departments.... Some academics will just be users. To use the tutorial programs is a user level skill. How to use them in teaching is an issue. That is a focus of our activity. That is what we have to try to do. The academic content of our [MEU] staff is crucial, because you have some chance of being able to convince academics if you are an academic. There can be a lot of misunderstanding between academics and technocrats. You need to speak the same language.

As the university endeavours to manage the investment in educational technology and provides funding through central, non-research mechanisms, then the demand for central or university funded support increases. Except for those universities that have strong machine bureaucratic elements which provide extensive central support, traditional support mechanisms in other universities such as academic staff development, computer and network support, audio visual services, computer training, and computer laboratories, are scattered across structural boundaries, and there are few mechanisms or processes to facilitate or encourage cooperation. This is most evident in provision of training for technology literacy and teaching method and support for courseware development. The latter now requires the support of a range of skilled staff including instructional designers, graphic designers, computer and web programmers, cognitive scientists and project managers. The use of multi-disciplinary teams to manage the increasingly complex systems development environment has long been recognised elsewhere as a solution. Inevitably tension between the different elements of the organisation grow because resource sharing and collaborative team work is not the common model for university work.

While IT in higher education has been a concern for some time, the current external budget drivers are having a significant impact on the timing and urgency of the need to ensure that educational technologies are relevant, cost effective, integrated with other university activities and/or provide a competitive edge.

The challenge for universities is to find the balance for the need to be efficient and effective without significantly changing the underlying professional roles of teaching and research. In professional bureaucracies, major innovation depends on cooperation and cross disciplinary activity which is not a mode of working with which many academic staff are familiar or comfortable, and there is therefore a hesitancy to change. Mintzberg (1979) states that 'As long as the environment remains stable, the Professional Bureaucracy encounters no problem. .... But dynamic conditions call for change ... that calls for another structural configuration.'

With respect to the structure of IT units, universities should consider the lessons learned from restructuring in industry, as well as how other professional organisations have incorporated technology into their core business. Recent studies of IT alignment within organisations have revealed that how companies organise IT activities may have a significant impact on the overall effectiveness of IT support (Blanton, Watson & Moody 1992). For example, in environments with a high degree of complexity, change, and uncertainty, organisations may improve the effectiveness of IT support by implementing IT organisational structures with higher states of both differentiation and integration. This means that in organisations like universities, with differentiated structures, IT is managed differently depending on user needs in various areas, rather than being standardised across the institution. Because of this, it also requires strong coordinating mechanisms to integrate across these different units. It is also proposed that effective integration mechanisms are those that promote feedback on IT performance, gain cross-functional participation in IT planning, and facilitate communication among IT groups and between IT groups and user groups through liaison functions. This is particularly relevant to the field of developing and managing educational technology, which requires considerable coordination across multiple stakeholders.

Furthermore, the effectiveness of IT unit structure is to some extent contingent upon the characteristics of the wider organisational strategy and structure. For example, Hodgkinson found, that 'a highly decentralised IT management style will likely be inappropriate in a company whose strategy emphasises increasing integration and coordination of business operations. Conversely, an excessively centralised IT management style will be inappropriate in a company with fiercely independent, autonomous business units' (1992, p.173). As we have seen in the context of this study, universities vary in the extent to which they emphasise more centralised integration and coordination versus decentralisation and faculty autonomy. Of course, IT can to some extent facilitate a hybrid, 'centrally decentralised' IT structure (von Simson 1990). For example:

At UNSW we have devolved the Academic IT support to faculties through service level agreements designed to match limited resources to faculty priority and to coordinate service through a centralised Help Desk. The feedback we now have is very positive and instead of not wanting anything to do with us eighteen months ago, most faculty are now continuing their IT support through us.

(site visit)

We discuss these issues concerning IT management structures in universities in more detail below.

Models of Organisation Structure

From our analysis of IT management in universities, we identify two general types of IT management structures - the professional and the machine bureaucratic - and discuss their strengths and weaknesses. Each of these 'ideal type' models exemplifies a specific trade-off between efficiency and effectiveness. While no one university is exactly represented by these ideal types, they serve to highlight how different forms of management can be successful, and furthermore, how combining the models within one organisation can create significant misfit. Pressures for the imposition of both models in universities do exist, so in that sense they may be seen as competing forms of fit. Forces for centralisation and standardisation compete with forces for devolution and autonomy. For example, in the Australian university system, the government imposes considerable regulation and administrative control which must be managed centrally and bureaucratically in the university. This contrasts with the self regulating focus of the professionals on their particular areas of teaching and research in the academic 'divisions'. The administrative and academic divisions typically have quite different priorities regarding their IT needs.

Later, we describe a commonly recommended solution - the federal IT structure - which is often claimed to best manage the competing demands of a centralised, cost-focused approach and a devolved, innovation-focused approach to IT management. The limitations of the federal structure in the university context are discussed and alternative forms are suggested, derived from industry examples or the case analysis.

IT Management in the Professional Bureaucracy

The older and usually larger traditional universities are characterised by a professional bureaucratic structure, the key characteristics of which are high levels of autonomy and discretion granted to the highly trained academics in the 'operating core' of the organisation. This very decentralised structure is primarily coordinated through the standardisation of professional skills, which is established and maintained by the professional associations which control the body of knowledge, and through collegial rather than strictly hierarchical management processes (Mintzberg 1979). In this traditional professional organisation structure, the managers of the more bureaucratic administrative support structures typically have only limited hierarchical control, and so they rely more on negotiated solutions than imposition of authority with regard to the management of professional operations. Thus management decision making is often the product of a plethora of committees, in which the primary coordinating mechanism is by mutual adjustment and horizontal liaison, rather than by hierarchical authority.

IT management in such organisations tends to be distributed and devolved, with a focus on supporting innovation in the individual faculties and departments. Central IT units in these universities manage the network infrastructure and administrative systems, and provide support to the academic units. Advantages of such an IT structure include the fostering of bottom-up innovation and a focus on the particular needs of users in the academic units. Potential disadvantages associated with the model include lack of cost control, and a proliferation of incompatible systems.

IT Management in the Machine Bureaucracy

The newer, often smaller universities also have a number of elements of the professional bureaucracy in common with the older universities. They are, however, typically characterised by a more centralised and bureaucratic structure. The strong traditions of faculty and academic autonomy found in the older universities are not so pronounced in these newer institutions. In this sense they have elements in common with machine bureaucratic structures. In the classic machine bureaucracy, the aim is to efficiently provide a routine service, with management and appraisal processes based on conformity with formalised procedures and budgets, the management emphasis is on hierarchical control, and roles are standardised. While few universities would identify with this ideal type model, in a number of them, the machine bureaucratic administrative element strongly competes with the professional academic elements.

Because of this stronger bureaucratic element, IT management in such organisations is more centralised, with a focus on obtaining cost efficiencies from more standardised operations and tight control of IT expenditures in the academic units. This central control of IT ensures a university-wide adoption of approved infrastructure, standard applications and software, thus minimising waste and duplication. It also makes inter-operability and integration across the university easier to achieve.

These two forms of IT management thus have quite different sets of strategies, structures, management processes, technologies and roles and skills. That is, they form distinctive configurations or gestalts of interdependent organisational elements. Trying to combine individual elements from across the two configurations can create a complex misfit. (For examples of such competing forms of fit from other industries see Johnston & Yetton 1996 and Yetton & Johnston 1996). Table 9 lists the differences between the two configurations, using the elements of the MIT90s framework.

Table 4: Characteristics of Two IT Configurations

Characteristics	Machine Bureaucracy	Professional Bureaucracy
Strategy	cost efficiency focus technology-driven	value added, innovation focususer-driven
Structure	centralised, hierarchical bureaucratic authority	decentralised, collegial expert authority
Management Processes	formal standards & procedures control emphasis position-based rewards distant from user	more flexible mgt processes project mgt, empowerment performance-based rewards close user-IT liaison
Systems	single dominant platform common IT standards common administrative systems	multiple platforms incompatible systems local academic systems
Roles/Skills	administrative focus long-serving staff	professional focus mobile staff

(adapted from Johnston & Yetton 1996)

To illustrate visually how the professional and machine bureaucratic types of IT organisations form distinctive competing gestalts, and to show the potential mismatch between an administrative and academic focus in IT management, we combine the models in Figure 5. What is obvious here is how the two forms actually pull the IT management in two different directions: one towards a value-added focus on client services, the other towards efficiency and integration across the institution.

Figure 5: Two Forms of IT Management

(adapted from Yetton & Johnston 1996)

Consistent with the trend in other areas of business, the emphasis in IT management in recent years has been on trying to obtain the benefits of both models without their disadvantages. That is, there has been a demand for low cost, innovative IT systems - both effective and efficient. But this generates a natural tension since the low cost tends to be associated with standardisation and centralisation, and innovation tends to be associated with decentralisation and business unit control.

Recent attempts to resolve these apparently inconsistent demands in large complex organisations have focused on notions of a federal IT management structure (see Figure 5). This represents one of the most elegant and complete designs for IT management and is advocated by many consultants and academics (Davenport, Eccles & Prusak 1992; von Simson 1990; Zmud, Boynton & Jacobs 1986).

The federal design involves creating a 'division of powers' in IT management between the corporate centre, business groups (e.g. faculties) and business units (e.g. departments/schools) in a typically divisionalised university structure. The IT management structure is essentially folded around this divisional design. Architectures and standards are controlled at corporate level, and user-driven development is located at the business group or unit level. IT services such as data operations, maintenance and telecommunications (service delivery) are managed separately from business system development and corporate IT management. They are treated as a 'shared resource' and managed as a quasi-business with complex transfer pricing systems. In effect, then, the current solution to the problem is to seek a balance between centralisation and devolution: to ask which activities and functions should be centralised and which devolved. Unfortunately, however, this theoretically elegant solution often does not work well in practice.

Figure 6: Current IT Management - the Federal Solution

(Source: Yetton, 1997)

On the surface, this 'strategic alignment' in a federal model seems to provide a solution to the tug-of-war between IT strategies that focus on cost efficiencies and corporate standards, and those that are value-added and business driven. However, the IT structure it creates depends on IT managers ensuring that the activities of the different work groups, which can all have strongly competing interests, are integrated across the whole organisation. This requires that the IT groups in individual academic units continue altruistically serving overall corporate goals and that the IT group at the centre remains entrepreneurial and focused on meeting user needs. The reality is that IT people at the unit level tend to be 'captured' by the goals of the academic units (or 'go feral', as one corporate IT manager put it), while IT people at the centre tend to become divorced from the business, and be seen as bureaucratically 'policing' a particular technology architecture and set of standards (Yetton 1997).

A structure which requires behaviours that are in such conflict with job pressures is usually under considerable stress. To make it work, you need IT managers with high interpersonal skills, particularly conflict management and negotiating skills, married to a good grasp of user needs. But these are usually in short supply. So this world is likely to be inherently unstable. It would be liable to revert to one of the two more stable structures: a centralised, cost-driven structure, or a decentralised, user-driven one. Alternatively, the organisation could find itself on an expensive 'reorganisation merry-go-round' as it oscillates between one structure and the other.

Alternative Models of IT Management

For the reasons outlined then, the federal system does not work well. Because of the nature of the technology and of universities' information systems, IT cannot easily be broken up and allocated separately to academic and administrative divisions. As a result, it is difficult for IT to become fully integrated into the university. So, if anything, it may be pushed further away from the core of business and wrapped around it. At best this is a second best solution in which IT will tend to be passive and reactive to the business of higher education rather than proactive and integral to it. Given these difficulties with the federal model, we proceed to tentatively explore two other emerging options for IT management structures: the divisional infrastructure model and the subsidiary model.

A Divisional Infrastructure Model

From our observation of IT management structures in many business organisations, we have observed in a few an emerging model which obtains the benefits of a centralised infrastructure without many of the disadvantages associated with strong corporate control of IT. In particular, this model obtains the cost benefits of a centralised IT infrastructure, without losing the benefits of user-driven IT innovation in the 'business units' of the organisation (or in universities, the academic divisions or faculties). There are two features underlying this emerging model - the shift from high cost systems proliferation to a common infrastructure, and the empowerment of users 'at the coalface' to build innovative systems for the delivery of service.

Usually, these have been seen as mutually exclusive solutions - centralised architectures and standards meant strict control of business unit IT, while devolved IT management meant that IT costs blew out as incompatible systems and duplicated data proliferated in the organisation. We have seen a few examples, however, where the appropriate application of an enabling infrastructure across the organisation was welcomed by semi-autonomous divisions, as it was shown to reduce their IT costs without constraining their current customer-focused IT systems. So, rather than corporate IT becoming a policeman for the architecture, the infrastructure, once in place, both maintained the users' current systems and thus their 'business unit' independence, as well as giving them options for new systems and enabling integration and inter-operability across the business units where required.

Applied to the university context, and viewing the academic faculties as semi-autonomous divisions, there could be significant advantage in the corporate centre of the university introducing such an enabling infrastructure, with the aim of simplifying the university IT platform and leveraging existing academic systems. University of New South Wales and University of Wollongong are potential examples of this model. Such an infrastructure could meet the demands for certain common administrative systems across the university, while supporting autonomous innovation around IT in the faculties. The challenge here is to leverage rather than constrain existing academic IT innovation. If successful, such a model would help move towards a world in which the benefits of innovation and cost control are obtained.

A Subsidiary Model

In some universities IT innovation for service delivery, especially in teaching and learning, has been managed by restructuring to create what is in effect a business unit or subsidiary separate from the mainstream operations of the university. These forms have two key characteristics. First, in these subsidiaries, IT is at the core of innovations in both the production and delivery of new modes of teaching and learning. Alternatively, they may be autonomous research centres in which IT innovation is central to their leading edge knowledge development. As we have seen in other organisations where IT is not separate from the core business (e.g. Yetton, Johnston & Craig 1994), IT is integral to the key operations concerning delivery of service to the customer. Second, and most importantly, IT in this model is 'owned' and mastered by the professionals themselves. Thus the professionals learn and master the changing technologies and apply them to their field of expertise (e.g. delivery of teaching and learning), rather than heavily relying on external technical specialists who sit between them, the technology, and the customer. Any important external technical expertise required is subjugated to the demands of the multi-disciplinary team of professionals (e.g. multi-skilled academics, teaching and learning experts, and IT specialists) building and delivering the service. These professionals also require a high level of technical literacy in addition to competence in their area of expertise. University of South Australia and Griffith University are examples of universities which have the potential to develop such a model.

Such a subsidiary model distinguishes IT management for innovation from that for standard university operations by structurally separating them. The advantage in this model is that it provides focused innovation, while not exposing the university to expensive and disruptive restructuring across the whole organisation. Where subsidiaries are successful, they can be grown as demand for their services increase. In some instances, the successful model can provide the basis for gradual learning and change across the whole institution. In the latter case, the university adopts a well-structured solution to the original complex problem but not capacity to solve the next complex problem. That capacity stays insulated within the subsidiary.

Conclusion

IT management structures in universities are differentiated according to their degree of centralisation or decentralisation, and according to the balance of professional and machine bureaucratic forces. Most have faculties working in a divisional framework - the extent of devolution of decision making and budgetary authority to those divisions is a key differentiating factor. The degree of convergence of structures among the corporate IT unit and other related areas also varies significantly across universities.

In universities' effort to use IT for innovative educational delivery, they should be mindful of Mintzberg's warning that the incentive for innovation in teaching and learning - which is traditionally weak in a professional bureaucracy - can be reduced by external controls. They should also be mindful that convergence and structural change are not a panacea. However, as Collier (1996) notes:

the ultimate formulation of a converged service is therefore a service model in which customer need determines service configuration, resources can be allocated flexibly across all areas of activity and wasteful duplication or competition avoided.... revised structures cannot of themselves guarantee these benefits, it is hard to see how they can be provided without management restructuring.

Although universities have some basic structural and cultural similarities, they are significantly divergent in operation and activity. Therefore, no one model will suit all. The experience of industry could be drawn upon to guide the changes to the IT service groups within organisations, and these need to adapt as technologies change. A major impediment to this realignment and change in work practice will be the sharp delineation between the industrial environment for different segments of the university workforce who will need to work very closely, preferably in multi-disciplinary teams. This has major implications for staff development. The subsequent re-skilling process will be a never-ending cycle.

Universities that are already adopting innovative IT management structures may be building a long term advantage as the environment of higher education changes to one in which IT plays a key role. These prime movers, if successful, build organisational learning and mastery around using IT in their core business, which makes it hard for others to imitate. Furthermore, the universities currently developing enabling infrastructures for IT could develop a cost advantage as well as a platform for innovation. However, this needs to be managed without constraining the academic divisions with too many machine bureaucratic controls, which tend to stifle local innovation.

Universities have the advantage of strong committee structures which, if used effectively, can overcome many of the barriers and resistance to the widespread adoption of educational technologies through effective consultation and co-ordination. Many universities have strengthened either their Teaching and Learning Committee, and/or their IT Strategic Planning Committee. There is also emerging evidence that outsourcing elements of IT support, training and courseware production is being considered, or has commenced in a limited way, at Australian universities (refer to Chapter 4 for further discussion on outsourcing). A significant number of universities have also followed the trend in the private sector to elevate the chief information officer into the senior executive group, in recognition of the increasingly important role of IT in the delivery and administration of higher education.