Asynchronous Learning Networks: Drexel's Experience
DR. STEPHEN J. ANDRIOLE, CIGNA Systems Philadelphia, Pa. RICHARD H. LYTLE, Director and CHARLTON A. MONSANTO, Associate Director Center for Multidisciplinary Information Systems Engineering Drexel University Philadelphia, Pa. In the fall of 1994 Drexel University began offering courses over an asynchronous learning network (ALN). While "unconventional" course and curriculum design and delivery is by no means new to the university, this was its first foray into interactive network-based "anytime-anyplace" education. The project is supported by the Alfred P. Sloan Foundation, which has an ongoing program in asynchronous learning. Drexel's mission is to explore the potential for asynchronous learning in the domain of information and software systems design for part-time and full-time students on and off campus, as well as students participating in Drexel's cooperative education program. Drexel's ALN uses Lotus Notes to support communications and problem-solving. Students access the ALN via Windows-based PCs and Macintoshes. Each student has Lotus Notes and a suite of cross-platform compatible design tools. Lotus Notes is a "groupware" tool that permits students to interact asynchronously-that is, anytime, anyplace via a technique known as "replication," which permits Lotus Notes users to get on a network, send comments and messages, and simultaneously receive comments and messages that were sent since the last time they logged onto the network. Five courses have been offered thus far, with numerous others in the pipeline. The early results are encouraging and, in spots, even surprising. There is-based on preliminary data-every reason to believe that ALN-based education can be extended into additional domains and that whole (especially professional) degree programs can be delivered over ALNs. Project Scope & Objectives Figure 1 shows the overall project structure. Objectives of the project include: Provide asynchronous learning opportunities for students on-campus, students in and outside the region on co-op; Demonstrate the integration of off-the-shelf computer-aided software engineering (CASE) tools into Lotus Notes to facilitate group learning and problem-solving; "Simulate" industrial information and software systems design on an ALN; Measure the impact of asynchronous learning across a broad spectrum of indicators ("student satisfaction," "retention," "degree completion time," "cost-effectiveness," etc.); and Implement self-pacing procedures. Information and software systems design and development is the domain. We have organized courses around the methods, tools and techniques that support the conversion of user and software requirements into maintainable information systems. The domain is "practice-oriented" and one from which the majority of students aspire to industrial careers. The student population includes part- and full-time students on campus, near campus and on co-op. Many of the undergraduates in the network have part-time jobs; most graduate students are employed full-time. The initial courses "converted" and delivered asynchronously include Systems Analysis & Design (4 times), Business Process Reengineering (twice), and Information Systems Policy & Administration (twice). Two other courses, User Computer Interaction and Information Systems Implementation, will be given shortly. The network is accessible to various kinds of students as well as the Lotus Educational Consortium-of which we are a member-and the University of Illinois, another Sloan grantee which hosts the Sloan ALN server that supports an on-going discussion of ALN issues, problems and challenges. A special feature of the project is its exploration of ALN-based self-pacing. As Figure 2 suggests, there are self-pacing opportunities for students that fall behind or proceed ahead of "average" students. This aspect of the project is important since ALN-based education (and training) should offer the same quality as conventional educational processes, but also provide some additional benefits. One may be self-pacing. Since the fall of 1994 we have made the course in Systems Analysis & Design "continuous," that is, we have offered the course term after term. This is significant because students that lag can stay in the course until they complete its requirements. Already students have taken incomplete grades in one term and completed the requirements in another. Self-pacing must work the other way as well. The ALN will also support accelerated learning as course sequences appear that will permit students to move faster through courses and thereby complete their degrees faster. This is especially important in a cooperative education environment where typical students take about five and a half years to earn their degrees. Accelerated learning could make a major difference in their personal and professional planning. The project is committed to evaluation. We are collecting data about "simple" and "complex" outcome measures. Early results-reported below-suggest that ALN-based education (and, by inference, training) is surprisingly cost-effective. Figure 7, near the end of this article, identifies the project's outcome measures. A questionnaire as well as data collection and analysis routines have been developed to determine over time the most effective ALN formulae. Interactive Asynchronous Learning Via Lotus Notes We are using Lotus Notes to orchestrate course activity. Lotus Notes is a groupware product from Lotus Development Corp. (and now IBM) that permits students and instructors to interact asynchronously. Students and instructors access the Lotus Notes server to transmit and receive discussion points and assignments. Each student, however, has his or her own course materials, applications programs and other data necessary to take the course on their individual client desktops or laptops. A key concept is "replication," which refers to the process by which students access the network to send and receive comments, messages, assignments and the like. When students access the ALN (via direct dialing into the network), the server immediately sends them all of the messages, comments and assignments that have been posted since the last time they logged on. The server also receives their input to the discussion or their assignments. As soon as this exchange is completed, Lotus Notes hangs up. The students then respond to the new information off-line, accessing the network again when they finish their work. This means that students need not spend much time at all actually connected to the ALN; instead, they are only actually on the network for a short period of time (usually five to ten minutes). Most of the work is done "locally" on their own computers. Figures 3 through 6 present screen displays from the ALN. These displays present the basics of ALN interaction. Most importantly, they convey a structure of ALN course delivery that is essential to success. Students and instructors must know an entire course's plan before the course begins. Students must be able to "see" a course in its entirety on the first day of class. It is also imperative that the learning environment be "user friendly." Students and instructors should find the system easy to use and intuitive. We have established some crude rules of thumb for assessing the usability of the ALN: ideally, for example, students should be able to learn how to use the network and its supporting applications programs in two hours or less. Results to Date The project began in January of 1994; by June of 1994 the ALN was up and running. The first courses were offered in the fall of 1994. Three courses have been converted to ALN form; the conversion of several others is underway. The methodology for the project is anchored in systems engineering, which calls for requirements modeling prior to design and development. We thus began with a set of educational objectives and then developed a course conversion/design methodology that permitted us to satisfy those objectives. The methodology requires that a set of "knowledge" and "ability" capabilities be achieved via the learning process. Those capabilities were in turn converted into a set of topics, sub-topics and assignments. Readings were then matched to the topics, sub-topics and assignments. We developed a set of tools and processes for monitoring ALN activity, a back-up server, an on-line grading system that permits instructors to grade student assignments as they are submitted, and integrated the cross-platform-compatible CASE tools into the network. Summary interaction data for the courses given to date appears below: # of Interactions for Approx. 8 Weeks 800 Interactions (for avge. course of 10 students) Preferred Interaction Times of Day 8 PM - 12 Midnight (36% of Total) 4 PM - 8 PM (22% of Total) 12 Midnight - 4 AM (14% of Total) So what have we learned about ALN-based course design and delivery? There are two kinds of observations. One set is based on a questionnaire administered after every course; the other is based on inferences made from questionnaire results, the experiences of the instructors, and empirical data about network behavior and patterns. Figure 7: Questionnaire results include: 80% would take another ALN course 85% felt they had more access to the instructor than in "conventional" course delivery 80% felt that conventional courses were more boring than the ALN course 75% did not miss in-class lectures 75% felt they had more communication with fellow students than in conventional courses 70% felt they learned more on the ALN-based course than they would have expected to learn in a conventional course 95% felt that seeing the ideas and assignments of others was useful While our sample data is still too small to offer verifiable generalizations, after several courses undeniable patterns are emerging. Students find the ALN more convenient and-much more importantly-a superb learning environment. The second set of results are inferential. It appears-after several course deliveries and questionnaire data analysis-that the following may well prove true: There is an enormous need for structure in an ALN environment, including: Opening and closing discussion windows Clear discussion topics, readings and assignments schedule Completely predictable course schedule All materials online and accessible Common course "look and feel," especially in a multiple course ALN Real-time monitoring of student performance It can be made very cost-effective: Hypothesized (maximum) quality ratios: - 1 instructor for every 30 students - 1 instructor + 1 ALN assistant per 50 students - 1 instructor + 2 ALN assistants per 75 students The methodology is repeatable The communications hardware and software is off-the-shelf and relatively inexpensive-and on the right price/performance trends Other conclusions: Industrial information and software systems design processes can be "simulated" via ALNs We can design a learning process and an interactive asynchronous learning environment that accommodates self-pacing CASE tools can be integrated into a groupware environment If ALN-based education and training is to succeed it must scale upwards cost-effectively. While we have worked to identify the educational critical success factors, there are a set of organizational conditions that are likely to predict ALN success. First-and perhaps foremost-we should recognize that ALN-based course and curriculum design and delivery is an educational (and training) specialization that not all faculty or institutions can-or will want to-"practice." It is also likely that faculty will-as always and ideally-be motivated to participate in the ALN educational and training process as part of their "normal duties," and not as an "overload." If ALN-based course design and delivery becomes an overload for faculty then it will become expensive and eventually end up competing with normal workload responsibilities. Institutions as a whole must also be committed to ALN-based education and training. If organizations regard the technology as a fad or as something in which they must become involved because of perceived competition, then they will not sustain ALNs as part of the primary delivery processes. The danger today is that asynchronous learning-along with other forms of "distance education"-will remain in the labs and in the hands of techno-educators-who seldom represent mainstream faculty interests. Plans Our plans include the conversion and delivery of additional courses, collection and analysis of additional data, implementation of full-scale self-pacing and refinement of an ALN "methodology" that can be transferred to other institutions. We also plan to extend ALN-based education and training to additional domains, and to eventually offer an entire professional degree program via the ALN. Over the next few years we plan to refine the course and curriculum design and delivery process, offer many more courses on the ALN, and continue to report our findings. Over time, we hope to contribute to the growing understanding of how alternative educational and training processes can be cost-effectively applied to a variety of learning challenges. Stephen J. Andriole, currently Chief Technology Officer and Senior Vice President for Technology Strategy at CIGNA Corp., was formerly a professor of Information Systems & Electrical and Computer Engineering and the Director of the Center for Multidisciplinary Information Systems Engineering (MISE Center) at Drexel University. E-mail:
[email protected] Richard H. Lytle is Dean of the College of Information Science & Technology at Drexel University and the current Director of the MISE Center. Charlton Monsanto is Associate Director of the MISE Center at Drexel.b The project described in this paper is supported by The Alfred P. Sloan Foundation; the Sloan Program Officer is Dr. A. Frank Mayadas. The Sloan Foundation has an ongoing research and development program in Asynchronous Learning Networks. We gratefully acknowledge the support of The Sloan Foundation, without whose support our work would not be possible. Products mentioned in this article: Lotus Notes; Lotus Development Corp., Cambridge, MA, (800) 343-5414