Home

Designing Distributed Learning Systems

by Sloan-C

Originally published in the Proceedings of the Orlando Multimedia 1997 Conference. Reproduced with the permission of Learning Technology Institute, 50 Culpeper St., Warrenton, VA 20186

ABSTRACT

Many companies are adopting an approach of hiring the skills they need, then discarding them and hiring others as needs change. Individuals must rapidly learn in their specialty and also develop new skill sets over the long-term. Access to intellectual capital and to learning and performance support systems is improving with developments in the World Wide Web. However access is not usability, and usability is not organizational impact. This paper provides principles and examples of distributed learning and support systems that can aid individual development. It considers motivation, speed of program development, and cost issues for some current projects where usability and impact are central design criteria.

I. Article

We are engaged in a major change in work and education. Global competition requires reductions in costs and increases in the rate of new technical knowledge and innovation. Guaranteed jobs are disappearing, and companies and individuals are often required to pay for training on new knowledge - or be replaced by competitors. Lifelong learning is the norm, with profound implications for the volume, speed, and quality of products and services required of those who support learning.

At this moment appears the Web -- a technology that enables worldwide competition among education providers. No longer must universities form around libraries because books are scarce and expensive and must be collected. Nor must the faculty be gathered around a physical resource in a library. The technology of the World Wide Web provides conferencing among widely dispersed individuals who are teaching and learning together, thus fostering competition between widely dispersed education providers. Now every learner at home or work can access a much less expensive and more readily available learning environment, which in some instances will be provided by the top "brands" of experts, companies, and schools. At some point, educators and trainers will be reducing costs, increasing quality, and increasing the rate of new knowledge and innovation about how to support learning - or disappearing themselves.

There are drawbacks to distributed learning: the Web can be addictive, so there can be a tendency to drop work and turn to the Web for distraction. Completion rates are typically very low for self-study courses. And distance education using video can become merely a replacement for standup lectures with little or no learner involvement.

We focus here on Web-based collaborative (vs. self-paced) learning, and in particular on asynchronous anytime, anyplace learning networks (ALN). We and others see ALN as one of the most promising forms of distributed learning systems. While synchronous video or recorded lectures can be provided on the Web, they do not generally support the interaction and collaboration that are the hallmarks of ALN. Nor does computer-based training (CBT) typically provide the interactions and collaboration with other learners that enrich and motivate ALN users. See the ALN Web page at http://www.aln.org for additional information.

Several universities and companies, including our Center for Innovation in Engineering Education, have been designing distributed learning systems. Some of these resemble traditional college courses with an online syllabus and lecture notes. Others include online simulations, collaborative projects among those at distant locations, and online submission of assignments. Some are more modular and support second shift workers in rural areas who cannot attend community colleges. Others are aimed at those in the community who have had little access to continuing education, who have few job skills, and now need to improve their employability.

In face-to-face courses with expected attendance, course content and evaluation standards can be presented orally and evolve during the course. In ALN courses these must be up front and explicit. Most of the principles for ALN are the same as for face-to-face courses, but become more important since learners may feel less invested.

Some principles and examples of distributed learning and support systems that can aid individual development are:

Adapt the learning system to the technical sophistication and tools of the intended audience. Simply providing access to the tools does not mean they will be usable by the learners. For example, individuals who lack a high school diploma, using older equipment, will generally need individual help at their own locations, to set up their systems. Rio Salado's program is an example of such work. A brief description is at http://www.sloan.org/education/trustee/rio4.html [7]. In contrast, graduate students in computer science will need little of such assistance.

Be clear and concise on what the assignments are, when they are due, who may work on them (individual or a team), and in what form they are to be submitted. Examples of submission forms are emailed text, a reply to a topic in a computer conferencing system, a computer file as an email attachment, or a paper document. See the Vanderbilt University School of Engineering EE274 course assignments page at http://jrbnt.vuse.vanderbilt.edu/274/ [9] for an example. Put the assignments all in one place. Include the number of points or other weighting assigned if the course is graded.

Make explicit the criteria for success: provide sample test problems, scoring criteria for projects with good examples from prior learners, and models of excellent problem solving. An example is a model position paper that defines the problem, examples of the problem, possible solutions, and a recommended solution. Another example is a sample of a programming test that includes one or more problems representative of each type that will appear on a test. A third example is a set of scoring criteria and examples for interpersonal communication skills that will be used in a role play. See the Course Details page at http://jrbnt.vuse.vanderbilt.edu/274/ [9] for an example. In the "Table of Contents for Introductory Materials" is a set of "Sample Final Exam Questions."

Get learners using what they learn. Examples are exercises, role plays, simulations, and projects. This involvement is one of the keys to learner investment of time and to course completion. Even role plays can be accomplished in a distributed learning system by having a partner at the terminal, with the software guiding and helping each partner score the other's performance [2]. The Vanderbilt University School of Engineering ES130 course "Laboratories" page provides an example of projects for freshmen. Look at http://www.vuse.vanderbilt.edu/es130 [10] for these sample activities.

Evaluate the learning environment to improve it (formative evaluation). Include at least an evaluation of learner reaction (the typical end of course rating) and specifics of learner performance in the class (e.g., scores on criterion measures that can be reliably graded). For an example ALN evaluation, go to the ALN Web homepage at http://www.aln.org [1] and in the left frame select the "Second International Conference on ALN." There look in the Conference Program. You will find a presentation by Roxanne Hiltz [5] that describes these issues. If possible follow up the course evaluation by tracking performance of course graduates in an organization where they use what they learned (e.g., ratings of supervisors, peers, and subordinates).

Motivation and Completion Rate

Distributed learning programs typically have low rates of completion. Correspondence courses, computer-based learning, web-based learning and other forms of self-paced instruction commonly suffer from this affliction. One reason for the popularity of video-based distance education is that it preserves scheduled classes and lectures with which students and faculty are already familiar, and that it provides expectations and motivation for regular learner engagement with the course (namely attendance at class lectures).

More people are likely to complete a distributed learning course that includes motivation and pacing aspects like the following, than will complete a course that does not:

• scheduled, graded assignments (at least once per week) with rapid feedback
• required contributions to online conferencing discussions
• a team that works together on a project
• a cohort of learners who progress through a course together
• a course grade, certification, or other indicator of value

Learners are motivated by many factors like a need to know or use the content, the value of certification or a degree to qualify for a job, or a desire for a high grade point average.

The goal is to increase the learners' engagement with the course. Basically that means their spending time on it. Frequent assignments with feedback, a sense that they can get their questions answered by another student or the facilitator, and a set of social expectations that come from being part of a group can help course completion.

Speed of program development

Program development is distinguished from delivery: a content expert may develop a program (with or without assistance on instruction design, graphics, programming and the like), then have facilitators work with learners. Speed of development depends on the normal factors like familiarity of the expert with the content, complexity of learner interactions (e.g., simulations vs. text), volume of content, and availability of development staff. One of the most important development speed issues is the type of interactions to be used. You can very rapidly transfer a syllabus and lecture notes contained in a word processor file to the Web. However setting up a conferencing system and developing automatically scored online exercises and secure learner access to their course grades can take considerable time for the first courses developed. Typically several months are required to develop a one semester college course. Shorter workshops can be developed by a skilled team with excellent tools in a few weeks. New Web authoring tools are likely to shorten the time or to improve quality and consistency. Examples of such tools are ToolBookII at http://www.asymetrix.com/ [3] and IBT Author at http://ibt.testprep.com/ [8]. While we are not specifically recommending these tools, they and others like them should be considered when you will be developing several courses or workshops.

Cost issues for some current projects where usability and impact are central design criteria

The goal here is to cut cost for equivalent or better learning. In very rough terms a 15 week academic course that already exists in lecture format can be converted in its simplest form for Web delivery for about $20K. This does not include expenditures for equipment and software. That is a very low figure in comparison to development of a new computer-based training program for business and professional use. Usability by learners must be factored into development costs, since learners who have very few skills and poor Web access will require much more support, and design sophistication for that support, than sophisticated users.

Development of new courses varies with the application. An academic course is likely to cost less than a professional course where complex technical skills and simulations are involved. We are just beginning to gain experience on costs for new ALN courses of various types. There is a tradeoff between comprehensive automated interactions (e.g., automatically graded simulations) and personnel costs to support the course. At one end is computer-based instruction that provides information and multiple choice questions, with no interaction with others. At the other is an individual tutorial coupled with collaborative projects and simulations. Visit Falcon Software at http://www.falconsoftware.com/falconweb/els.html to see ELS - an electronic laboratory simulator.

Looking Ahead

Design of distributed learning systems is evolving rapidly as those who broadly subsidize training and education struggle with increasing needs and escalating costs. The growth of the World Wide Web encourages competition and provides new tools to facilitate the work. We have suggested basic steps to move beyond access to usability and to organization impact. That strange mixture of competition and cooperation that is prevalent in other industries is coming to training and education, and we intend to foster it. We invite you to join us.

Acknowledgements

We gratefully acknowledge Frank Mayadas at the Alfred P. Sloan Foundation, and Arthur Brodersen, John Crocetti, Carolyn Gale, and Eric McMaster at the Center for Innovation in Engineering Education at Vanderbilt University for support and discussion of the research and evaluation on which this paper is based.

II. REFERENCES

1. ALN (Asynchronous Learning Networks) Web page: http://www.aln.org
2. Campbell, J. O., Lison, C. A., Borsook, T. K., Hoover, J. A., & Arnold, P. (1995). Using computer and video technologies to develop interpersonal skills. Computers In Human Behavior, 11(2), 223-239.
3. Asymetrix Toolbook II: http://www.asymetrix.com/
4. Electronic Laboratory Simulator (ELS) from Falcon Software: http://www.falconsoftware.com/falconweb/els.html
5. Hiltz, Roxanne (1995). Presentation on ALN and evaluation "Launching a Degree Program in Information Systems via Virtual Classroom," Second International Conference on ALN, Conference Program, Near Campus ALN: http://www.aln.org
6. Journal and Magazine of Asynchronous Learning Networks: http://www.aln.org
7. Rio Salado program: http://www.sloan.org/education/trustee/rio4.html
8. Stanford Testing Systems' IBT Author: http://ibt.testprep.com/
9. Vanderbilt University School of Engineering EE274 course assignments and final exam questions: http://jrbnt.vuse.vanderbilt.edu/274/
10. Vanderbilt University School of Engineering ES130 course "Laboratories" page: http://www.vuse.vanderbilt.edu/es130

III. Author Information

J. Olin Campbell is Research Associate Professor of Engineering Education at Vanderbilt University, and President of Campbell Associates. He can be reached at 4617 Chalmers Dr., Nashville, TN 37215; email: campbejo@ctrvax.vanderbilt.edu; home page: http://olinc.vuse.vanderbilt.edu/olininfo/ and phone at (615) 322-0110. His work focuses on strategies for assessment, learning, and performance support using computer simulation, and on strategies and tools to make development of distributed learning more cost effective. He is Associate Editor of the Web and Journal of Asynchronous Learning Networks (http://www.aln.org/). He holds a Ph.D. from Stanford in educational psychology, an M. Div in theology from Union Theological Seminary (New York), and a B. A. from Yale in psychology.

John R. Bourne is Professor of Electrical and Computer Engineering and Professor of the Management of Technology at Vanderbilt University. His email is john.bourne@vanderbilt.edu and his homepage can be found at http://www.vuse.vanderbilt.edu/~bournejr/persinfo.htm. He has been on the faculty at Vanderbilt for the past 28 years with sabbaticals in Sweden and Northern Telecom. He is Editor of the Web and Journal of Asynchronous Learning Networks (http://www.aln.org/) and is editor-in-chief of the Critical Reviews in Biomedical Engineering.