UAB STUDENT INFORMATION SYSTEM
David A. Corliss
This document describes the framework within which the SISMAC will operate. It begins with a summary of the Strategic Agenda goals and importance of the SIS to achieving those goals. It goes on the define the users of the system and what the users require of a successful SIS. The following sections discuss issues related to mainframe versus client server solutions, institutional readiness for change, and how we came to be where we are today. These sections conclude with the notion that we should think in client/server terms as guiding principles, if not a technical solution. The last section outlines the complex nature of the problems that we must deal with and makes recommendations on how we organize ourselves to meet the challenges before us.
Two specific goals relevant to undergraduate students resulted from several months of analysis and discussion: to increase retention between the first and second years and to increase graduation rates. To meet these goals several processes and their associated policies must be examined, restructured if necessary, and monitored on a continuous basis for effectiveness. These include
 | the recruitment of more students who are more likely to succeed |
| the identification of at-risk students during the admissions process |
| the identification of students' goals when they enter the university |
| the effective monitoring of students' progress and academic readiness |
| the structuring of curricula that allow students to make steady and timely progress toward their goals |
| the allocation of institutional resources in the most educationally sound manner |
The interrelations among these processes have been mapped out in detail and many initiatives designed to perform these processes more effectively have already begun. Among the key elements that are critical to the success of these initiatives is well organized and easily accessible information in the form of a computerized student information system (SIS). Information is the glue that binds the people and the processes together.
To appreciate the types and organization of the required information we must define the end users of the SIS. First and foremost are the students themselves. Students must know where they stand throughout their association with the university and what options are available to them at every step of the way. The information they need can be provided directly to them via internet and kiosk or indirectly through an advisor, a staff member, or a faculty member. In this view these latter users can be considered to be intermediate in spite of the fact that they will technically spend more time in front of the computer screen than the students. These users are, in addition, those who are responsible for gathering the information and ensuring its accuracy. A third group of users includes those who monitor outcomes and, in the ideal, make data-driven policy decisions based on available data. These too are end users.
These three user groups have some common information requirements. First, all the information about a student must be available from one common source. Every piece of data about a student from recruitment, through admission, enrollment, graduation, and beyond should be accessible to the other two groups of users (with proper security clearance, of course) at any time and from any place.
The necessity of having a common data source leads logically to the second information requirement: point-of-contact data entry and decision making. When a student interacts with a recruiter, an admissions officer, a financial aid officer, an advisor, any other appropriate staff person, a faculty member, or an administrator, that person must be able to call up that student's record, make any changes necessary, and have those changes immediately accessible throughout the system. These users should also be able to make immediate decisions based on up-to-date information.
This ability to make immediate decisions leads to the third information requirement: everyone must operate under the same set of rules related to students and programs. Policies regarding such things as admissions requirements, transfer credit articulation, satisfactory academic progress, course availability, pre-requisite/co-requisite checking, and graduation requirements must be embodied in a set of rules that are stored in the system and accessible to everyone at all times. Furthermore, when it is necessary to change them it should be possible to do so without extensive reprogramming. The people who are responsible for developing the rules should be able to change the rules within the restraints established by the system as a whole.
A fourth information requirement is the ability to easily define work flow and automate it using a standard messaging system. The intermediate groups of users should be automatically notified when tasks need to be completed and students should be notified of the deadlines they need to meet. The policy decision makers should have standard reports available on a regular basis and be notified of any tasks that they need to perform. Automation ensures that the right people do the right things at the right time.
The requirements of having an institution-wide database, point-of-contact data entry, an institution-wide sets of rules, and work flow automation represent a fundamental shift away from a centralized, hierarchical model to a more distributed model at both the organizational and information services (IS) levels. They also represent a shift from an IS-centered system to a user-centered system. These shifts enable greater flexibility for all users which, in turn, demands greater flexibility at the IS level.
Flexibility at the user levels means being able to respond to changes in the internal and external environments quickly and easily using data-driven decision making. It also means being able to adapt the system to the unique needs of the multiple schools and programs within the university. The responses to environmental changes and diversity need to be made without compromising the integrity of the institution-wide system.
In this user-centered model users have flexibility to the degree allowed by the constraints of the system. Thus, IS must be flexible enough to quickly and easily change the constraints of the system, if that is what is needed to respond to new user requirements. IS must also be able to respond to rapidly changing technology if the changes represent more efficient or cost effective ways of performing either user- and mission-critical functions. For IS to meet these challenges there are a number of general system requirements:
| Modularity-This means that the components of the system (e.g., databases, rules, user interface, documents) are separate and distinct so that changes can be made to a given component without affecting all the other existing components. For example, tables could be added to the database or fields to a table without requiring changes to functions that work with the existing data. Procedures that operate on the new data could then be added without changing existing procedures. Code for general procedures should be reusable. |
| Scalability-There should be measurable performance gains for any additions of hardware or changes in the operating system, the application software, or the work flow done on the system. This means that new users or application requirements can be accommodated more easily, thereby extending the usable life of the system. |
| Interoperability-This requirement is fundamental to the two listed above. At one level it means that the components of the system can be installed on different hardware platforms or operate on multiple platforms across many sites. At another level it means that some of the components are not vendor specific. The system could, for example, work with different databases rather than proprietary databases developed for a specific application. |
While the definitions of these three critical system requirements vary somewhat from one source to another in the computer literature, together they give IS the ability to respond to user demands and technology changes more rapidly. This ability results from three factors:
| Giving IS the programming and maintenance tools for rapid application development. |
| Distributing some of the ability to adapt the system to specific requirements to the users. |
| Moving the application from a single mainframe to multiple servers which can be replaced or augmented more easily and cost effectively than a mainframe. |
Client/server systems are, by definition, user-centered, distributed, modular, scalable, and interoperable. "…information can be easily distributed between client machines and server machines and yet be presented coherently without delay or without end-user knowledge of the network topology. Client machines in local area networks can be server machines, and server machines can be clients, interchangeably. Every machine in the network has client and server qualities." In other words, the network of computers becomes the computer.
There are many technical models of client/server systems but the essentials include a relational database, sets of business rules, and the client programs. In today's office the client is most often a desktop running Microsoft Windows and its associated office applications. One of the key advantages of properly designed client/server systems is that users can leverage their knowledge of these applications when learning and using a client/server SIS. The properly designed system uses the Windows standard menu structure, keyboard and mouse conventions, and the graphical interface. It also allows users to easily extract information from the database and import it into a variety of desktop applications for further processing. Form letters, for example, can be easily composed in their final form and mail merges done without intervention on the part of IS.
The very things that make a client/server solution do not come without costs, in the broadest sense of the word. Several issues must be addressed when considering moving to such a solution:
| Client/server solutions are more complex by virtue of their modularity, scalability, interoperability, and the distributed nature of their environment It seems to be a fundamental rule of computing that, the more user friendly a program or system becomes, the more complex it becomes on the backend. The demands on IS thus increase. |
| Client/server solutions are more complex by virtue of their demand for user participation. "…business rules, and particularly their exceptions, are really political issues, not technical issues. Technology cannot now, nor will it ever be able to, replace the political process that is necessary for making decisions or coming to consensus on the business rules (or their exceptions) in an organization." |
| The institutional dollar costs per user for initial implementation and long term maintenance are higher than they are for mainframe systems. Because of the user-centered nature of client/server systems, the costs may be distributed over many more units within the institution, however. |
| Because of the above it is often financially cheaper to adapt and continue to use mainframe solutions because of the cost of translating legacy systems into client/server systems and the associated learning curve for all concerned.2 The financial cost savings may be short term, however. The long term costs of sustaining a mainframe solution include the relative difficulty of taking advantage of those things that the make the user-centered, client/server model attractive and continued reliance on expensive mainframe computers for any expansion (though costs are coming down). |
To succeed in changing the SIS to meet the stated goals, there must be a match between the level of complexity of the implementation and the ability of the university to handle that degree of complexity at both the technical and political levels. In the summer of 1996 consultants from IBM conducted a study of institutional readiness. They examined both the technical and political climate and were able to compare the characteristics of UAB with other organizations that had failed and succeeded with the implementation of client/server systems.
The report affirmed that UAB has a very high probability of success. They concluded that the technical infrastructure, with some enhancements, was capable of sustaining a client/server SIS. They also concluded that users, administration, and IS were sufficiently committed to a new system. The strategic planning process and the organization of various working groups and committees had established a strong foundation for change even before the study was conducted.
The university was ready to take on the challenge of implementing a client/server SIS and contracted with TRG in early 1996 to do so. The product was still in development at the time the contract was signed, but the potential benefits of the system seemed worth the risk, as it did to several other universities that had signed on early in the process. In the end, TRG could not deliver and backed out of its contract. UAB recovered its investment, regrouped, and examined other companies that were also developing client/server solutions. After considering the options provided by these companies, and the alternative of continuing in-house development, the SIS Oversight Committee recommended that we contract with PeopleSoft to implement a client/server system that they were developing. Once again, the principles embodied in such a system were compelling enough to outweigh the inherent risks. Furthermore, PeopleSoft is a company with a solid track record of implementing client/server systems on large scales in the human resources and financial areas of management.
There were two problems with the recommendation. First, the PeopleSoft product was more expensive to begin with than that of TRG, though not excessively. Second, there were reports from institutions implementing other client/server solutions that unbudgeted implementation costs could amount to as much as twice the contracted price of the system. Considerable consulting expertise is apparently required to reengineer the business processes to the software. Although UAB was well ahead of the game in examining its practices, and was deemed ready for such an undertaking, the magnitude of these additional costs was too much of an unknown to justify the risk.
Another factor that needed to be considered was the fact that, had we contracted with PeopleSoft, the actual implementation date would have been very close to the year 2000. Given the inevitable delays in software implementation, the institution could not take the risk of not fixing the year-2000 problem in its current system. Even now, we need data that carry the year-2000 stamp. Thus, considerable resources needed to be devoted to modifying the current STARS system. Six of the eight Computer Center programmers are now working on this task.
Given the decisions that have been made, we are faced with the challenge of enhancing our current SIS to meet the user and system requirements outlined above to the extent possible. One of the things that we learned in talking with other institutions during the brief TRG contract is that the UAB SIS is far ahead of many others. We are thus building on a good solid foundation. If we assume that a client/server solution is as much a way of thinking about people and processes as it is a technology, we should be able to enhance the system in important ways.
Our mantra should be, "First the student, then ourselves." Or, to paraphrase James Carville, "It's the student, stupid." We must bear this in mind as we consider enhancements to the system and the ongoing integration and implementation of policies that directly affect students. Our success will be measured by how well we help staff and faculty help students identify and achieve their individual goals.
The role of the SISMAC is to take a strategic point of view. It is to integrate the people and processes around a set of stored data and to turn those data into information that are useful to the students and to the staff and faculty who work with the students. The best way to organize ourselves is around the people, processes, and information that help the students.
To take the strategic point of view we must confront our complexity to see how all the parts fit together. We must keep the big picture in mind at all times.
We are a complex institution because of the variety of different students that we admit to our many different programs. A third of the undergraduate student body consists of the so-called traditional students who start college directly out of high school. The rest of the undergraduate student body consists of transfer students. There is a large population of graduate students enrolled in programs at both ends of campus. There are multiple undergraduate and first professional programs in the traditional disciplines, like business and engineering, and in health care. Each has unique characteristics and requirements but they share common decision points, processes, and information requirements.
Figure 1 illustrates how the Strategic Agenda working group that dealt with student issues attempted to simplify the problems it faced. It shows the decision points and processes affecting undergraduate students. The basic elements will be similar for graduate and first professional students. Each decision point (the diamonds) in this diagram embodies policies that need to be coded as rules within the system and triggered automatically based on the data available. Each rectangular shape represents a process or the outcome of a decision. Every single element in the diagram depends in some way on data and rules that should be incorporated into a SIS.
Figure 1. Undergraduate student decision points, processes, and outcomes.
There are various components of the current SIS that are in place, under development, or in the proposal phase that deal with various elements of the flow diagram. Each involves managing considerable detail. Rather than have the committee as a whole involved in all the details, it is essential that members of the committee take responsibility for various components. This means forming and working with users of the SIS who have front-line responsibility for using these components. The following working groups are a preliminary recommendation of how these groups should be constituted. It is subject to change pending further details.
| Recruiting |
| Application Processing |
| Financial Aid |
| Placement Testing |
| Articulation |
| Advising |
| Registration |
| Student Accounting |
| Degree Audit |
| Course Scheduling |
| Resource Management |
| Outcomes assessment |
| Kiosk and Web Access |
Changes to the SIS proposed by the user groups will fall into two categories: those that are mandatory, like the federal regulations that affect financial aid, and those that are elective. It is the elective changes that the SISMAC must consider from its strategic point of view. We will also be expected to consider the SIS implications of policy changes that come from other committees around campus and work closely with these committees to see that they are implemented properly.
I have described the framework within which the SISMAC will operate, our goals, and some ways of thinking about the SIS that should enable us to enhance the functionality of the current system. I suspect that we will find ourselves encountering questions of process as much as questions of technology. Finally, I have recommended an approach to dealing with the complexity of the system that expands the role of the users of the system by creating small focus groups. Our success will be measured by the success of these users in their roles helping students to meet their goals.