Debunking the distance-learning myth
We've overcome the challenges of electronic distribution, and no technical barriers remain to prevent distance learning from becoming an overnight reality. So, why aren't more companies taking advantage of it?
By Nicholas Cravotta, Technical Editor -- EDN, January 9, 2003
|
The complexity of many engineering tools has made effective training an important check box on the list of essential criteria when evaluating new products. Live training, however, is expensive, often requiring travel and a significant amount of unbroken blocks of time. It can also be hard to justify spending even a couple of hundred dollars to learn about a tool that you think you may, but aren't positive, you might want to use later in your career.
For decades, distance learning has been put forth as a low-cost panacea for the inherent difficulties in transferring information among people, whether in an academic or a corporate environment and whether transferring technical knowledge or policy. Such learning can reduce or eliminate travel costs and free individuals to study at their own convenience and pace. One classic application of distance learning is to provide a means for disseminating "tribal knowledge" within a company. Another is to allow people to download evaluation tools and tutorials from the Web at low or no cost, an especially appealing proposition to out-of-work engineers who want to hone their skills between job interviews. From a company perspective, an engineer trained to use a tool is more likely to recommend its purchase when such a tool is required. Additionally, properly trained users are expected to place a lighter load on support channels.
Distance learning has met the first major hurdle—distribution—and connection bandwidth provides acceptable presentation, even of audio and limited video, in a practical, pervasive, and cost-effective manner. So, why aren't more companies taking advantage of distance learning as an educational or marketing tool? As appealing as the possibilities are, significantly large hurdles still exist for distance-learning technology to overcome.
Content is king
Several considerations of distance learning arise beyond distribution. These considerations include content, interactivity, automation of processes, personal touch, and cost. Quality of content is the most critical of these factors. A variety of methods and corresponding tools exists for capturing content, but good tools don't guarantee quality presentations. Distance learning requires more preparation than simply putting a training manual online. In many cases, the quality of printed material decreases when you post it online because it becomes more difficult to use efficiently; compare adding bookmarks and notes to a printed catalog with locating and downloading the same material each time you need to access it.
Distance learning has two major forms: asynchronous, meaning that you can access it at any time and in any place, and synchronous, meaning that you can access it in real time. Synchronous teaching is similar to in-person teaching in that a real person uses support materials to give a presentation, can dynamically alter the material to meet student needs, and may be able to immediately answer questions. The mistake that users make with asynchronous teaching is thinking that it differs little from synchronous teaching: There is no live person giving the presentation (it could be prerecorded) or available to answer questions, so the support materials must stand on their own. Designers of self-paced materials must keep self-pacing in mind when developing them.
A major vendor investing in distance learning gives the following estimates in distance learning: It takes 25 to 35 hours to prepare one hour of material for a live presentation and 150 to 200 hours for one hour of an asynchronous presentation. This scenario presents a five- to 10-times increase in the time-to-cost ratio. In other words, to break even on time alone, you need to be able to use the exact same materials for five to 10 sessions. The time-to-cost ratio increases as you add multimedia elements or video.
It is rare to get material down perfectly the first time. The live lecturer can incorporate feedback from students by changing the presentation the next time that he or she gives it. Simple changes are relativity easy with a live presentation. For example, you can reorder material by just placing a few arrows on your notes. However, these changes become more complex to manage and execute with prepared—that is, static—materials. For material that changes over time, giving live presentations is often more time-efficient in the long run.
Some of the best tools for capturing content are fairly straightforward to use. Presidia, for example, offers a plug-in to PowerPoint that times animations, synchronizes audio files, and converts presentations to a flash format that you can play from a standard Web browser. The simple addition of audio plays a critical role in involving the viewer and personalizing an otherwise-dull presentation. Camtasia Studio from TechSmith is a tool for creating flash video of captured screen activity that complements Presidia's offering. For example, you can illustrate the use of a tool by executing the process on your computer and record your actions to show someone else what you did and what the result was. You can edit captured activity; add audio; and add other highlighting features, such as arrows or hyperlinks. Camtasia Studio provides a similar effect to having WebEx host a Net meeting for you at www.webex.com. The difference is that WebEx is a live, synchronous replay of screen capture that allows viewers to talk via phone conference, and Camtasia Studio is a recorded, asynchronous capture. Used together, slides with audio and links to illustrated processes can effectively demonstrate complex concepts.
Touch it to make it yours
Simply placing material on a computer screen, however, is not teaching. The main challenge of distance learning is that, because students retain more information when they are personally involved, the material has to go beyond mere text. Multimedia presentations help increase retention but at a higher production cost, and they tend to provide a more passive type of interaction.
The best presentations have conditional aspects that allow students to interact with material and test results for themselves. For example, a presentation on resistors would allow a student to create configurations of resistors using different values. This approach allows students to test their understanding and explore cases that may be obvious to an instructor but troublesome for a student. Such a presentation would be similar to posing a question to a live instructor—an action that is simple for a live person but would require extensive programming of a simulation environment even for simple online problems. Canned presentations don't easily offer such interaction and cannot answer questions that no one anticipated or previously asked. To fill such "holes" in presentations, students still require access to a live expert. Distance-learning instructors can adapt presentations over time to address more commonly asked questions but at continuing cost.
One of the touted benefits of distance learning is that it frees up an instructor's time. However, this "benefit" is a myth. If anything, distance learning, although usually eliminating in-person feedback, creates many alternative feedback channels, such as e-mail, forums, and online surveys, that many students feel more comfortable using exactly because of their more impersonal nature. Additionally, administrative factors increase because distance learning means that instructors must manage passwords and access issues. These issues complicate the process over and above simply signing up for and attending class. With asynchronous materials, many company officials mistakenly believe that they can open the virtual doors of their classrooms to an unlimited number of students. However, as the number of students served increases, the feedback load also increases, creating serious scaling issues (see sidebar "Forum-scalability issues").
One method of addressing the increased flood of feedback and questions is the use of a online forum. Instead of using a private channel to speak to an expert, students post questions in a common forum for all interested parties to view. Broadcasting questions may prevent several students from asking the same question, thereby increasing the diversity of questions a leader can answer. A forum also creates an environment for student interaction and dialogue, a critical element of learning; in posing a question or giving an answer, a student must exercise his or her knowledge of the material. Additionally—and the most desirable theoretical timesaver of all—students can answer each other's questions.
Several forum models are in development today across the Internet. Newsgroups offer a member-managed forum but are public and difficult to control. For example, newsgroups need to manage troublemakers who push users to buy their new book or get free cell phones. They also need to manage confidential material and do housekeeping, such as removing old messages and building FAQ (frequently asked questions) files. Forum services are available from companies such as Yahoo, but control is still an issue. Private forums are available from companies such as Web Crossing, offering the sponsoring company complete control of material and membership with additional management functions, albeit at a higher cost than "free" alternatives.
The dynamic of student participation in which students help each other learn and reduce the load on the expert is a goal of the online DSP forums that Texas Instruments created. With its online forum, the company intended not to create another customer-support channel, because those channels already exist, but to enable engineers to help other engineers. Such forums still require some oversight, such as when an engineer answers a question poorly or inaccurately, meaning that the sponsoring vendor might get the blame for not catching and posting a correction. However, the cost of providing such oversight is potentially less than the value that the community of engineers gains. A key problem is consolidating information from other channels, such as e-mail to individuals, training portals, info centers, and discussion groups. As the number of users grows, you can no longer count on one person to handle all channels, and you need a mechanism in place to share information to prevent duplication of work. Additionally, some channels have mechanisms for tracking customers and problems; consider that a problem someone raises in a forum doesn't receive the same attention or tracking as a call to customer support.
Promoting forum participation is tricky, as is any attempt to try to make people behave a certain way. If contributors have no incentives to post, users may find too few postings or responses to make the forum worth visiting. To give members incentives to post, forums could tie the course grade to posting, limit the number of questions a user can ask based on the number of questions the user has helped answer, or offer perks. Web Crossing, for example, offers a "virtual-dollars" concept, in which you earn dollars based on either how much you participate or how high others rank your postings. You use your dollars to "vote" for improvements you want to see in the next version of the software; this system appears to be effective for getting feedback from the people who use the software and have good ideas for improving it. Unfortunately, incentive-based posting often results in low-quality postings: Users end up viewing quantity—not quality—as important, and worthwhile postings become lost in a sea of otherwise-less-useful material. Measuring the quality of postings is feasible for a small group but not for large groups. For example, even checking whether someone duplicated someone else's posting—not to mention verifying whether the posting is relevant to other nearby postings—becomes a near-Herculean task.
Automation
A successful forum results in a large number of postings. Theoretically, these postings store a great deal of valuable information, either as observations or well-stated questions and responses. One proposed benefit of forums is that they are a self-generating knowledge base that people can consult when they have a question. Having an instructor repeatedly answer the same question is not an ideal use of time. Collecting questions and answers into a knowledge base promises to free up an instructor's time by letting students query the knowledge base before they disturb a human.
The first problem to overcome is creating a base that is large enough to answer a reasonable number of questions, thus making it worth consulting. However, as postings accumulate, the challenge of finding postings that are relevant to the question at hand becomes more burdensome. Monitoring forums is a costly proposition, because monitoring goes far beyond simply removing inappropriate postings to the point of organizing and ranking posts.
Certainly, good questions and good responses to them are candidates for a knowledge base. Extracting this information proves to yet be one of the great challenges of distance learning. It can easily take an hour to process one good question and response, as you generalize the question to widen its scope, edit and rephrase ambiguities, pass to others for review, assign keywords for the search engine, and post it to the base. Simply putting all postings into the knowledge base degrades the quality of the base and increases the time it takes to find an appropriate answer. If the searching process is often fruitless, people simply won't use it. For example, try finding an answer on the Microsoft knowledge base, which mixes products, versions, and terms, making finding an answer to a simple question tedious and time-consuming. Once you find a good question, you need to create search vectors to the knowledge it contains. You also have to accommodate the myriad ways in which people pose effectively the same question using different wording. For example, a vendor describes how it took a handful of developers several months to create the knowledge base for a single chapter in a biology textbook. Useful knowledge bases are expensive to build, and some are short-lived, because content changes substantially and regularly, such as with new versions of tools. Such knowledge bases are not worth the effort, because developers will not have finished building them when people need them most and the most savings are possible.
Note that a successful knowledge base effectively transfers relevant information. Such a base does not necessarily reduce the number of queries you receive. First, you have to convince and teach people to access the knowledge base before they try to contact a person. More important, however, a good knowledge base increases the overall expertise of your students—but at a price. Texas Instruments, for example, finds that, although the number of support calls has somewhat decreased as the knowledge base has grown, the complexity of the questions users are asking has increased to the point that those initially answering the questions are often no longer able to.
The personal touch
One major resistance to distance learning is that no one wants to learn from a computer. The more human a presentation appears, the more information students tend to retain, and the longer they appear interested in the material. Many people perceive learning from a computer as inferior to live learning in a classroom. Distance learning gives students more control over learning, but this control is a plus only if the student is an active learner. Many students forget that you can learn from a book; they approach learning as a "you-teach-me" relationship, a passive process in which an expert pours knowledge into someone's head. Distance learning does not solve the problem of motivating students. In contrast, many key traits of distance learning aggravate these issues. However, distance learning does allow those who are motivated to learn faster and more efficiently.
Distance learning does not make sense for small, individualized courses. You simply can't get the economies of scale that justify creating custom material. You also need to account for the lifetime of a student's membership: Are people involved for years or for a few weeks? This factor affects how many times you can reuse materials before you have to revise them. The more students that you serve, the more sense it makes to use distance learning. Note that, as the number of students increases, so does the required quality of the content: Pruning a corporate-policy presentation by two minutes multiplied by 10,000 employees adds up to a lot of saved dollars.
In one approach to distance learning, some companies put evaluation tools with demos on the Web that anyone can download. However, good reasons exist for not taking this tack. The chief danger is that garage engineers will overrun your support channels, preventing you from serving the people who haven't yet committed to placing an order but most likely will if they can properly evaluate a tool. It's the trade-off between getting a tool into as many hands as possible and not having people learning on "your dime."
A final challenge for distance learning is to determine whether individual programs are successful. Success is a difficult metric to define. Simply getting lots of customer postings may eat support resources and yield no sales or savings in other support departments. Additionally, some benefits may be immeasurable: Many more people view postings than make them. If you have no way of tracking who uses materials and how they use them, you have no idea how useful those materials are and which are worth the cost of maintaining. The purpose of distance learning is to pass on information in a more efficient and less costly manner than other means allow. You need a clearly defined way to determine whether you have met this goal.
Distance learning is an interesting case study because it reveals many of the problems associated with bandwidth-based businesses. Many business models focus solely on providing consumers with bandwidth or information channels, claiming that content will drive the use of that bandwidth. However, these companies often fail to account for the cost of producing viable content. For example, producing an hour of video for a potential audience of 1000 people is unlikely to make financial sense or create much more than a "blip" of bandwidth demand in the overall scheme of things. A general perception is that distance learning should be less expensive than live learning, but that perception is often wrong.
Distance learning is not a panacea. However, when you use it as a supplement to traditional methods, it can significantly reduce costs, enhance comprehension, and shift the burden of teaching from experts back onto students. For example, companies could present a four-day course as two days of online background, in which viewers write their questions and bring them to the two days of live discussion, reducing student travel time by 50% and enabling a speaker to teach twice as many sessions. This approach maximizes the use of in-person time, satisfies the students' need for access to an expert who can answer their questions, and elevates the discussion.
The story of distance learning isn't that it can do everything or that it is problem- free. In reality, it brings a new set of problems to the learning table. However, it enables us to do much more than we could before.
| Acknowledgments | ||
| Special thanks to Mentor Graphics, Texas Instruments, and Americ Azevedo from Goldwarp for their contributions to this article. | ||
Acknowledgments
Special thanks to Mentor Graphics, Texas Instruments, and Americ Azevedo from Goldwarp for their contributions to this article.
| Author Information |
 Nicholas Cravotta is communications technical editor for EDN. He teaches a programming course at the University of California—Berkeley using distance-learning technology and may be the first instructor there not to come within 100 miles of campus during his class. You can reach him at 1-530-346-8556, fax 1-530346-9777, or nick@edn.com. |
For more information...
When you contact any of the following manufacturers directly, please let them know you read about them in EDN.
Centra
www.centra.com
Docent
www.docent.com
eHelp Corp
www.ehelp.com
Goldwarp
www.goldwarp.com
Microsoft
www.microsoft.com
Presedia
www.presedia.com
TechSmith
www.techsmith.com
Texas Instruments
www.ti.com
Web Crossing
www.webcrossing.com
WebEx
www.webex.com
Yahoo
www.yahoo.com
