Designing Effective Learning Environments for Web-Delivery

Stanley Trollip, PhD

Capella University


Throughout the world, people are creating both instruction and learning environments on the World Wide Web. These range from short episodes that are Web-based versions of traditional computer-based instruction or training (CBI or CBT) to months-long courses that offer university credit. Both types are proliferating at a great pace, and both have good and bad exemplars. The issue that this paper addresses is how to make Web-based learning effective, no matter what format it is.

The single most important thing to remember is that if you are planning to create learning environments for the Web, you need to start the process at the beginning, even if you have lots of material from existing traditional courses. This does not mean that you will have to discard what you already have, only that you may have to discard any preconceived notions of how your existing assets will fit into the Web-based course. In this paper, I will describe a process that works for all types of Web environment and illustrate it with a case study from Capella University.

Learner goals and audience analysis

To begin at the beginning means that you have to define what you want your learners to learn, as well as understanding who your learners are and what environment they operate within. This latter part is a traditional audience analysis. (Many of the details of the process described in this paper are given in greater detail in Alessi and Trollip (2001), particularly in Part 3.)

In the case of Capella University, each course has its own content goals, which have to be established by the faculty teams responsible for curricular development. Currently, there are about 450 courses, so the range of content goals is very diverse. Although we would like to be in a position in which being on the Web does not define our content goals, the current constraints of the Web does impact what can be covered. For example, bandwidth considerations preclude many of our learners from being able to view streaming video or run Web-based simulations and other similar applications. This means we have to be very careful about selecting content that depends on such media. Fortunately, there are ways around these constraints in most situations.

Aside from issues surrounding content, there are other very important aspects of our audience that have to be taken into account. First, the average age of our graduate learners is nearly 45 years old. By definition, they also hold at least a Bachelor’s degree. Second, most of them are working professionals. Third, most have families. Fourth, they live in all 50 states in the United States and in about 40 other countries. Fifth, about half of them access the Web through America OnLine (AOL) and a large percentage (perhaps 30%) have 28.8 kbs modems. Only 10% have high-speed connections greater than 56.6 kbs. Sixth, about 15% of the learners do not have a computer capable of running multimedia applications (most notably sound). Finally, all are taking courses for university graduate credit.

One other important item at the beginning of the process is determining the financial aspects of course delivery. Someone has to decide how much can be spent on each course. As with the audience, the range of course costs can be very broad. Some higher education institutions say that they are spending about a million US dollars per course. Others, spend only a few thousand. How much can be invested in each course depends on how much money and other resources are available, such as personnel time, access to media production, availability of programmers, and so on.


Once you have determined the audience attributes and goals and you know what your budget is, you have to decide how they impact your design process to build and deliver your Web-based courses. Each analysis will yield a different set of designs, so there is no one correct way. However, if you follow the process correctly, your outcomes are likely to be successful.

Each of the audience attributes influences the decisions that have to be made in formulating the learning environment to build to deliver the courses. For example, if your audience analysis revealed that most potential learners were Spanish speaking and only spoke English as a second language, the reading level of the materials you developed would likely have to be lower than for native English speakers. Or if many in your target audience were poor typists, you may have to leave out exercises that required extended input. Or if your learners did not have ready access to computers, you may have to decide not to use Web delivery in favor of some other means to which all had access.

Let us now look at the implications of each of Capella’s audience attributes defined above.

Average age is 45; holds a Bachelor’s degree. These two attributes have profound implications, the most important one of which is that the learners are adults and will want to be treated as such. They are not teenagers who are happy to be told what to do and learn. Typically, adult learners know what they want and will not tolerate being treated as children.

From an instructional perspective, this likely means that they will not be happy being lectured to. That is, any attempt to transform traditional classroom lectures into a Web format is likely to be unsuccessful. For some faculty members, this is a difficult transition because they are used to being the "fount of knowledge" or "sage on the stage." In this learning environment with these learners, the instructor role has to be more of a facilitator (or "guide on the side"), guiding the learners through the course. For some, this is a loss of prestige, and they feel their importance is diminsihed.

Adult learners also place a great deal of importance on relevance. They are less tolerant of having to spend time on activities or courses that do not help them directly. From a design perspective, this means that most activities in a course must result in knowledge or skills that the learners can take back to their workplace as quickly as possible.

One other item from adult learning theory that has to be considered in the course design is convenience. Adults are typically busy and greatly appreciate every effort to make their lives less complicated. Web-based learning environments are intrinsically convenient, so this is almost a built-in benefit.

Working professionals. The fact that Capella learners are working professionals means that the course design has to ensure content relevance, as mentioned above. In addition, because the learners have a great deal of experience it is important for them to feel valued. That is, they want to contribute their experiences and share what they have learned on the job with the other learners. This means that the course design has to accommodate this sharing.

One other implications of the learners being working professionals is that they do not want, if possible, to have to drive through heavy traffic at the end of the day in order to attend classes. Not only are they tired, but they prefer to go home. Once again, the issue of convenience plays a big role.

Learners have families. The implication of this is simple. Learners would prefer to go home after work so they can spend time with their families. They also do not want to spend much time on weekends away from home.

Learners are scattered all over the world. There are several implications of having learners all over the world. First, because of the difference in time zones, any synchronous activities are going to pose difficulties for some learners. Therefore, it is better from a design perspective to have largely asynchronous courses.

Second, one of the most important design issues in this situation is to build courses that take the distance out of distance education. That is, the course design must encourage community, collaboration, cooperation, and closeness. As soon as learners who are at a distance feel that they are isolated, they are more likely to drop out of a course. So every effort must be made to make learners feel wanted. (In reality, good faculty practices also play a very important role in minimizing distance.)

AOL access and slow modems. Both of these issues mean that many learners have little bandwidth with which to play streaming media, download large files, and so on. From a design perspective, this means that courses need to be largely text-based, except that media-rich pieces may have to be sent to learners via CD-ROM or videotape – both of which add to the logistical complexities of running a course.

15% of learners cannot play sound. If you accept this state of affairs, your design will have to leave out any multimedia. At Capella, we decided that this was too great a constraint and required all learners to have multimedia computers.

Graduate credit. Since all courses are being taken for graduate credit implies that learners have to both put in the amount of time typically expected of a graduate course, and be able to demonstrate that they can perform at a graduate level.


Finalizing the design

As can be seen from the brief discussion above of the major attributes of the target population, very specific design decisions have to be made in order to accommodate the preferences and styles of the learners. Sometimes, specific attributes may demand contradictory design features. For example, synchronous activities, such as chat rooms, certainly help break down the barriers of distance. However, they are also very inconvenient. As a designer, you have to make a choice between these opposites by determining which offers the greater benefit for learners.

It is also at the design phase that you need to start factoring in the budgetary issues. The type of course you would develop if you had $500,000 for it would be very different from the one for which you had $50,000, and different again from the one developed for $5,000. (It is important to note that the whole budget and project management processes are extremely important, but will not be covered in any detail here. Alessi and Trollip (2001) has extensive coverage of both.)

In the case of Capella University graduate courses, the following design was decided upon and implemented.

First, for convenience, courses are all Web-based and built upon an asynchronous model. That is, within reasonable constraints, such as weekly units and a 12-week course, learners can make their contributions at their own convenience.

Second, to accommodate slow bandwidth, courses are largely text-based and rely minimally on media other than graphics. Although streaming audio is acceptable, learners are adamant that they are not interested in listening unless the content contributes substantively to the course. They do not want audio for audio’s sake.

Third, in order to draw upon the rich experience brought to the course by the working professionals who are the learners, the core of the course is an asynchronous threaded discussion group that encourages learners to both share what they have learned over the years and to help each other. This has been extremely successful, and some learners form virtual networks that carry on long after the course is finished.

Fourth, to build community and to reduce distance, every learner is required to post a substantial response to two or three questions posed by the instructor each week. In addition, each learner must respond to the postings of at least two other learners. This creates the beginnings of a dialog that often blossoms into a set of interactions that exceed those found in most traditional classrooms. With a dozen to fifteen learners in a class, I have often had 150 – 175 postings per week in the discussion group. Although some are short, such as "Good idea" or "That is interesting," most contributions have substance. We are currently experimenting with a slight change to this process, whereby instead of just responding to two postings, learners must now also ask a question about the content. We think this will improve learning by requiring deeper processing, as well as improve the dialog and interaction.

One other design feature that we are beginning to build into courses is mandatory collaborative projects. Not only does this help build community, but also provides invaluable practice for learners for their workplaces. More and more, we are told by employers that virtual teambuilding skills are very valuable.

Fifth, to provide relevance, we encourage learners to focus most projects on issues at work. This both helps them articulate issues that are important to them, and often results in them getting feedback and advice from other learners who have experienced similar situations.

Sixth, Capella courses typically require learners to spend about 8 to 12 hours per week, reading, responding, or working on projects. These projects form the second major basis for evaluating learners (the first being participation in the discussion group).

Seventh, and this is not a design issue but rather an implementation item, the success of even an well-designed course will depend to a large extent on the quality of the instructor. Online learners demand that their instructors are present – that is, they must be participating in the course at all times. It is usually not good enough for an instructor to make a few comments here and there once or twice a week. Adult learners will generally not tolerate this and may well drop out (or contact the university administration to complain).

At Capella University, all potential instructors are required to be learners in a faculty development course (online, of course) so that they may learn the techniques of being an online instructor and experience being a learner in the same environment. Such faculty development is essential for successful online courses, as we have found to our dismay when untrained instructors have tried to go online, often with poor results.



The discussion above is intended to highlight the process of designing an effective Web-based learning environment. The Capella example serves only to show how the process is applied. The Capella model may not work if any of the underlying audience attributes are different, if the instructional goals vary, or if the budget is not the same.

I am confident, however, that the process – a typical instructional design process – would work if the goal were an instructorless, self-standing piece of instruction that was to last an hour or so. It would also work if the Web-based environment were an adjunct to a traditional classroom. What is important is that a clear process is followed that is built around a clear understanding of the target audience and the intended instructional goals.



The book below contains many references pertinent to the discussion of this paper.