Associate Professor Steve Hansen, School of Computing & Information Technology, University of Western Sydney,Campbelltown Campus, PO Box 555
Campbelltown, NSW 2560 Australia, s.hansen@uws.edu.au
This paper presents both theoretical and case study work being currently carried out at the University of Western Sydney in providing an integrated online tutorial registration system. This system is integrated with the timetable system and is accessed by students and staff through an integrated teaching support infrastructure called PlatformWeb. The timetable and tutorial registration system is decentralised in the administration and management with interfaces to faculty support staff and by access at the subject coordinator level. During the year 2000, there were over 65,000 online registrations from more than 9,000 individual students across 861 subjects.
With the acceptance and availability of web interfaces to an institution's information systems, it is now possible to provide a variety of integrated services via the web to both staff and students. These services covering traditional administration tasks such as student enrolment details to providing the student/subject data for online teaching delivery. In line with this, a project called PlatformWeb was started in 1998, at the University of Western Sydney, to provide an integrated information infrastructure to support and enhance web based online teaching activities. The design and specifications for this were developed from applying a modeling of staff behaviour taken from Rogers (Rogers, 1995) on the diffusion and adoption of innovations and an institution's decision making processes for innovations. Following surveying of staff, a user-centric design was adopted that linked the enrolment, timetable, subject and course, staff and student enrolment administration systems with an internally developed online teaching delivery module. The system providing a 'one-stop-shop' interface for staff and students. The project was developed in-house and by the end of 1999 had grown to be over 30,000 line of code, was used for online teaching support by over 350 staff, and used by almost all teaching staff for associated functions (such as access to subject lists). By this stage, more than 85% of all students had at least one online subject and except for 200-300 students, the total student population of the Macarthur campus had now accessed the system.
At the end of 1999, it was decided to integrate the existing timetable system with a to-be-developed online tutorial registration system to provide integrated online administration/management with online student/staff access. Up to this time, various methods for tutorial registrations were in use, from prescriptive timetables given out by faculties, some LAN and web based allocations for collections of subjects, lists on doors, various preferential allocation systems to the 'come and we'll fix it up on the day' approach. There was a high level of dissatisfaction with these methods, most having high management and associated running costs (both time and money). Student acceptance of a web based online tutorial registration system was not seen as a problem since the students (about 8,000) had accepted the web based PlatformWeb infrastructure. However, the adoption and acceptance across a whole campus of another 'new' system by the academic and administration support staff was seen to be a potential problem in line with their past experiences. Accordingly, the design and specifications for the tutorial registration system were developed from applying the same modeling of behaviour, taken from Rogers (Rogers, 1995) as used for the PlatformWeb project. That is, a 'user-centric' and 'bottom-up' design and adoption model. This model is outlined in the following sections. Along with this model, the advantages of decentralising the management of the registration system so that the management and administrative load was taken off the central administration and distributed down to the user through the faculty support and academic staff was seen to be possible with this form of design process.
The Rogers model has the innovation-decision process being made up of phases. These phases being Knowledge, Persuasion, Decision, Implementation and Confirmation. According to this descriptive model, potential adopters of an innovation must:-
In addition, the uptake of an innovation maybe influenced by the inherent 'innovativeness' of the individual or organisational unit. The prediction being that individuals/organisations who are predisposed to being innovative will adapt an innovation earlier than those who are less predisposed. On one extreme of the distribution are the 'Innovators'. These are the risk takers and pioneers who adopt an innovation very early in the diffusion process. On the other extreme are the 'Laggards' who resist adopting an innovation until rather late in the diffusion process, if ever. The diffusion process typically following an 'S' curve as shown below.
Figure 1. The adoption of innovations. The vertical axis represents numbers, the horizontal time. Initially a few 'innovators' are involved, then the bulk. One of the objectives of the tutorial registration project was to develop an implementation that would compress the time scale.
Overview of Adopter and Developer Models.These models view the diffusion/adoption process from the viewpoint of a potential developer of an innovation, in terms of the interaction with the 'clients'. Basically, this leads to either 'developer-based' or 'adopter-based' (Surry and Farquhar, 1997).
In the developer-based model, the assumption is that a 'top-down' approach of a developer with a supposedly 'superior' technological product will increase the diffusion, along with the assumption that potential adopters are viewed as predisposed to adopt innovations that are quantifiably superior. These models are usually based on either an organisation or the developer choosing the 'authority innovation decision' process to 'force' and innovation. The assumption being that the client will either through the power/authority social structure adopt or will 'intuitively' recognise the relative advantages of the innovation.
In contrast, the adopter-based models, such as from Andrews and Goodson (1991), focus on the human, social and interpersonal aspects of innovation diffusion. Adopter-based models view the end user, as the individual who will ultimately implement the innovation in a practical setting, as the primary force for change. In particular, these models reject the assumption that superior product and practices will automatically be attractive to potential adopters.
In particular, researchers such as Burkman (Surry and Farquhar, 1997) have developed particular development models based on the 'individual', which can be readily applied to a collective and indeed to an organisation when taken as an 'individual unit'. The Burkman model emphasises the needs and perceptions of the potential adopters as being the primary forces that influence adoption. This model consists of 5 steps each of which is concerned about the characteristics of the individual adopter:
In pragmatic terms, these user-centric models when applied to the adoption of web based delivery, predict that for all the advantages the adoption theoretically can bring and regardless of the technical superior of any particular product, the primary force for adoption is the end users, that is the staff and students involved.
The above approach was taken in the PlatformWeb project with surveying and interaction with the academic staff with great success. Applying this same approach to the design and implementation of the tutorial registration system identified the administration timetabling support staff as the crucial 'adopters' with the students as secondary ones.
A series of workshops was organised with the administration staff to jointly determine their perceived needs and to involve them in the 'ownership' of the project. From these sessions, a set of specifications and a web interface was developed and modified by further discussions with the staff. These are given in the following sections in this paper. The basic 'philosophy' of the system, whether it should be preferential or other was developed by considering 'economic' issues and ease of use by the student. This is also set out in the following sections of the paper.
On trialing the system in the end of 1999, the operation was far better than anticipated, with full and enthusiastic endorsement from staff and students. This prompted another look at the process, with some rather interesting results. From this further research work on the human decision making process has been initiated.
Since the initial project started, this system has now been implemented across other campuses of the University of Western Sydney (UWS was a federated University with three members, but from 2001 has been fully integrated into a single structure). From the start of 2001, the whole university will be using this system.
In considering the operation of the tutorial registration system, whether it should be preferential, or first come-first serve , or other arrangements, a number of issues were considered. These were based on simplicity of operation and an effort to minimise the management and administration components. With preferential systems, although there may be advantages in scheduling, there are a number of management issues such as the initial setting up, the collection of the data, the management of 'late comers' and 'changers'. In addition, from previous preferential systems the satisfaction rate by the students had never been very high. In particular, in preferential systems, the strength of a preference is not assessed, that is two users may rank choice C as their second choice, but this may not be a strong preference for one (ranking as third choice may be just as good) but may be mandatory for another. The result tending to ignore the absolute necessities of some users and the corresponding lack of necessity of others, by 'averaging' out the pain among all users. The result is that there is general lack of satisfaction, with additional management in dealing with those with complaints.
In previous prescriptive systems, where complete (or partially complete) placements are determined by administration and then given out to the user, although fairly 'successful' in new student enrolments (as judged by the complaints etc), the management of the initial scheduling and the management arising from undesirable combinations from the bulk of the student population was also seen to be uneconomic.
In line with involving the user as much as possible, it was decided to try out a simple 'first-come-first-serve' system, with the user determining their own combinations (with limits set by the system), and in their own time after the initial enabling date. The management of such a system being seen to be the minimum, with the main components being the initial scheduling.
At the Macarthur Campus, the scheduling of teaching activities (ie tutorial registration system), is a joint effort of faculty timetable support staff, with an administration timetable officer using a scheduling program called 'Admin'. At the other campuses there are similar arrangements, but using the package 'Syllabus Plus'. The front end design of this system takes the output from 'Admin' and 'Syllabus Plus' and after parsing for inconsistencies producing a common database. From then on, the operation is independent of the scheduling package.
In order to produce the actual timetable, additional data is needed on top of the scheduling. This includes allocating comments to a subject or to an activity, setting or clearing 'full' flags, allocating staff to a subject or an activity, grouping activities together for various reasons, setting other constraints not entered into the scheduling packages, enabling a subject or groups of subjects for online registrations and attaching maximum registration numbers. The last item is usually different from the numbers needed for the scheduling program as a number of 'spare' places may be allocated to cater for non-web entered registrations.
These operations are all provided though a web interface available to the timetable support staff at the faculty level, with their changes being real-time reflected in the online timetable.
The online timetable also displays the status of registrations for each teaching activity, that is the number of places left.
On logging into PlatformWeb, students are presented with their current enrolment (links automatically become active if that subject uses a web based teaching delivery package, to the package), and a menu item is available to the tutorial registration module.
On following the link, all subjects the student has enrolled for the year are displayed grouped by teaching sessions. For each session, all these subjects are displayed, as with the online timetable, along with a simple web interface for selecting activities. The support staff can set various restraints such as course restrictions and home campus restrictions that when enabled only display to students 'allowable' activities for registration (this is discussed further later in the paper along with how well students make choices).
Students may change their selections in any one subject up to 4 times, after this they are directed to the faculty.
A weekly planner is also available to show their program
The web interface available for the support staff, gives various reports on registration and enrolment numbers, tutorial and subject student lists, allows the enabling of full flags (before the activity is actually full), movement of students from activity to activity (and off the subject and manual entry onto the subject).
The actual operation and web interfaces were developed from an interactive set of workshops with the timetable support staff. A block diagrams of the operation is given in figures 2 and 3 with some sample screen shots in the following figures.
Figure 2. Overview of the elements of the registration system
Figure 3. The operation of the registration process from the student view
Figure 4. Display on logging in, listing enrolled subjects and links to tutorial registration module.
Figure 5. Typical timetable listing of subjects and activities
From the allocation of subject coordinators to the timetable, subjects are associated with staff, on academic staff logging in, they are given a listing of their subjects. The web interface for academic staff is similar to the administration support staff, but only allows access to the setting or clearing of the full flag (this is set automatically when an activity is full, but it may be set to give spare spots for manual allocations), moving students between activities (including onto/off the system) and reports such as lists of students.
The operation is decentralised down to the support staff in the faculties and to the academic subject coordinator. The main student administration having no management functions other than the subject enrollment and subjects on offer (as generated in the timetable preparation).
The decisions to enable a given subject and enable given activities in a subject are decentralised down to the faculty level and faculties were also able to determine the degree of usage themselves. The web interfaces providing the necessary data and business logic to make this workable. In actual operation, almost all subjects were enabled.
The usage over 2000 is given in table1.
| Period 2000 |
Subjects | Individual Students | Online registrations | Comments |
| Semester 1 Macarthur |
350 | 6632 | 27,587 | Not all subjects online |
| Semester 2 Macarthur |
362 | 6175 | 24,371 | Not all subjects online |
| Semester 3 Hawkesbury |
129 | 2204 | 12,748 | Two faculities only |
| Other (MBAs, Summer & Winter Semesters at Macarthur) |
20 | 394 | 491 | |
| Totals over 2000 | 861 | More than 9,000 | 65197 | |
Table 1. Usage of the online tutorial registration system over 2000
Students were sent out a simple letter informing them of the basic operation of the system, how and where to log on and what to do for web access. Computer labs at the university were made available (for continuing students) and manual entry was still possible. For new, first time enrolling students at the start of the year, different faculties adopted different approaches. These included using a prescriptive timetable for these students, arranging computer access on their orientation, and to just let the students register as for the continuing students. The prescriptive non-web method produced the most problems which was solved by entering the data onto the web system, and then allowing students then to make their own changes. For the faculty who had arranged computer access on orientation day, only 12 students (out of over 600) attended. Unknown to the faculty, the how and where tutorial registration information had been inadvertently sent to these students and the bulk of students had already used the system as was the case with the remaining faculties.
The acceptance of the system, as determined by formal feedback and workshop discussions, was very high, with the 'complaint rate' dropping to almost zero! In addition, the number of changes by students of their choices was extremely low (see figure 7). The other surprising aspect was that the anticipated problems of resource allocation flowing from a first-come-first-serve allocation to do with resource clashes and efficient allocations did not arise. This prompted further investigation into the student selection process.
The following aspects were identified:-
From discussions with psychology staff with experience in human decision making, these results fit various models of behaviour based on the perceptions of 'control' people have when making decisions. Basically, the proposition being that the greater the perception of control, the more the individual takes ownership of their decisions. This model of behaviour was consistent over the year 2000.
Following the implementation at the start of 2000, the system was used for each teaching period throughout the remainder of 2000, with a number of Faculties of the Hawkesbury member of UWS also going online for the second half of the year.
Figure 6 shows the registration pattern across the Hawkesbury and Macarthur members for the second semester in 2000. Both campuses had (different), well defined dates for registration, but the Macarthur students had prior experience with the system. As is shown, about 35-45% of students register in the first day. These figures give an indication of system load and computer hardware/software requirements to run the registration system.
Figure 6. The registration of students after the enabling date. Two campuses are compared, with the students having different experience of the system.
Figure 7. The proportion of students making changes to their initial selections.
There have been many positive outcomes from using this online registration system. Its integration with an existing administration/teaching delivery web infrastructure has strengthened and complemented the whole.
There has been a cost savings from the decrease of time being spent by administrative staff in tutorial allocations administration. In particular, with the previous systems, the support staff had peak load times at the start of teaching periods, and high stress levels in dealing with irate students (and academic staff!). There has also been a tightening of the data integrity of the student data from the feedback of the online access. By using the web interfaces that complemented the existing systems, the resource planning exercise was enhanced, with better communication between staff and system and easier access for its management.
The decentralising of the management and administration has also resulted in better and more efficient operation as the data quality is improved the closer it is to the 'coal face'. Through this, much more of the local knowledge of the tutorial and timetable implementation has been captured and incorporated into the system. Coupled with this has been the increased participation by academic staff in the smooth operation of their subjects due to their perception of greater ownership of the processes.
The first-come-first-serve operation as compared to a preferential operation had many advantages. Examples being, with a preferential system, the strength of preference is not indicated by a the choice, resulting with 'unsatisfactory' allocations on preference, with the online first-come-first-serve it was found that the 'eager beavers' registered first and were happy, late comers, knowing they were responsible for being late (and it seems that the web interface enhances the perception of an individual being in control of the decisions), making appropriate choices without complaints to the institution.
Issues of student access and equity of access to the web, have been shown to be minor, the number of students with web access problems being small, manual entries into the system from student contact with the support staff being satisfactory.
Finally, from general comments from students and staff, the movement to students exercising more control and responsibility over their programs of study, has improved the university's 'image' and enhanced the whole learning environment.
Andrews, D., & Goodson, L.(1991). A Comparative Analysis of Models of Instructional Design. In G.J Anglin (Ed), Instructional Technology: Past, Present, and Future, Englewood, CO:Libraries Unlimited, 102-116
Hansen, S., Deshpande, Y., and Murugesan, S., (1999). Adoption of Web-Based Teaching Delivery by Staff In Educational Institutions: Issues, Strategies and a Pilot Study, Proc of the Australian Web Conference 99 (AusWeb99), 379 –396.
Rogers, E., (1995). Diffusion of Innovations, 4th Edition, New York: The Free Press
Surry, D., and Farquhar, J., Diffusion Theory and Instructional Technology. Journal of Instructional Science and Technology, Vol 2, No 1, ISSN: 1324-0781 at http://www.usq.edu.au/electpub/e-jist/vol2no1/article2.htm
Hypertext reference for PlatformWeb http://platformweb.uws.edu.au/platformweb
Steve Hansen, © 2001. The authors assign to Southern Cross University and other educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The authors also grant a non-exclusive licence to Southern Cross University to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the conference papers and for the document to be published on mirrors on the World Wide Web.