Usability-tested e-learning? not until the market requires it

When DigitalThink's design team was developing training software for Circuit City, they did something unusual. Team members donned the polo shirts of Circuit City sales associates and spent a couple of days working the sales floor, selling Palm Pilots, cameras, and stereos. Then they repaired to their lab to begin designing 200 one-hour training courses for sales associates.

Rolling user-centered design techniques like task analysis into e-learning software development is pretty uncommon on both the corporate training and higher education fronts. "A lot of people think they can't afford usability," says Michael Korkuska, managing director of design strategy at DigitalThink, an e-learning software development firm. DigitalThink does usability testing for every type of product offering, rather than on a more expensive project-by-project basis. "It's not clear the market cares about quality so much as price and speed improvements," says Korkuska. "Until the market demands it, and is willing to pay, it won't get it."

The idea of e-learning usability is still so new it's barely on the market's radar screen. Usability testing traditionally offers a way to ensure a product or a piece of software "works"—that is, that most people can make sense of it and use it with ease. But here's the rub: Established sets of principles of what "works" for online learning, based on research findings or industry best practices, do not yet exist. "There are very few tests that show e-learning is doing anything," says Jay Cross, founder of the Internet Time Group, a consulting firm. "There's no way for most buyers to make any evaluation except to say it looks pretty."

The usability standards and tests that have proven their worthiness for e-commerce and other applications need to be tweaked a bit before being applied to e-learning—tools for streamlining a site so someone can book a flight with ease aren't directly applicable to making an online biology class navigable and engaging, for example. For e-learning, "usability is not the major issue: learnability is," says Don Norman, a professor of computer science at Northwestern University, eLearn magazine Editorial Advisory Board member, and cofounder of the Nielsen Norman Group, an executive consulting firm geared toward human-centered product and service design. "But the very same practices followed by the User Interface (UI) community can be used to ensure learnability: learner-centered, iterative design, and frequent tests."

When the market eventually insists on usability testing, e-learning usability experts will be the ones making sure e-learning courseware is as effective as it is stylish—and the data they uncover may offer insights on the high dropout rates from online education courses. First, experts would test software in a lab, assessing how test subjects handle learning tools or cope with lots of data. They'd keep testing once the software was in use, assessing how new students navigated virtual classrooms, for example. Students might be asked to rate their classroom on its ease of use. Without the demand for such practices, vendors ignore UI issues in their rush to release products, according to Cross. "This will not change until vendors see there is a competitive advantage to doing this right," he says.

Educators like Joseph Konstan, an associate professor of computer science and engineering at the University of Minnesota, suggest there is a competitive advantage to designing e-learning software with users in mind. In a recent column wirtten for eLearn, Konstan wrote:

My message to developers of distance learning and instructional computing technologies is simple: Follow the user-centered design practices that human-computer interaction specialists teach. Follow me around for a semester to see how important it is to have groupwork and direct observation. Watch how a lecturer reads the class as the class is listening and learning. Collect the artifacts that make up a class. Then, don't sacrifice any of these capabilities until you can propose something that surpasses them. Teaching more isn't teaching better.

This is a tall order, and it's not surprising most of the models developed for e-learning have essentially taken the easy road. "What people have historically done is taken books formatted traditionally and put them online," says Carol Vallone, CEO and president of WebCT, an elearning technology provider for higher education.

Recent developments in digital imaging, streaming audio and video, and interactive interfaces, along with bandwidth increases, have potential to make e-learning software more dynamic, but vendors should be careful about being wooed by fancy technology, however impressive, said participants at an E-learning Special Interest Group (SIG) discussion at the CHI (Computer-Human Interaction) Conference in Seattle last April. In addition to usability standards, participants suggested, pedagogical guidelines are needed to help determine whether that slick Flash animation embedded into courseware helps students learn or alienates them.

The participants also noted the potential for games or simulations to enrich e-learning courseware, and wondered whether e-learning developers can learn from the power of games to simultaneously impart skills and entertain. "Simulations are powerful learning tools," says Don Norman. After all, users are fully engaged by a well-designed game, navigating the interface seamlessly after learning basic skills, eager to learn whatever it takes to progress to the next level of play—all while having fun. Sounds like a user-centered e-learning software developer's dream—imagine college students approaching their Calculus III class this way!

The problem with games is they are expensive to develop. Commercial games for the Sony, Nintendo, and Microsoft platforms cost millions of dollars per game to develop. "Education simply doesn't have those kinds of funds to spend," Norman says. "It is a crime that the game industry can devote more time, effort, and creative skill to their pursuits than can the educational industry."

User-centered designers have an upward battle ahead, both in proving to vendors, and the education market, that usability is worth the money, time, and effort, and in developing standards of e-learning usability that can be shared by developers everywhere. But reading usability expert Ben Shneiderman's description of an effective computer system, which appears in his book Designing the User Interface: Strategies for Effective Human-Computer Interaction, makes the battle seem worthwhile, and is an inspiration for those developing e-learning courseware over the next few years: "Well designed, effective computer systems generate positive feelings of success, competence, mastery, and clarity in the user community. When an interactive system is well-designed, the interface almost disappears, enabling users to concentrate on their work, exploration, or pleasure."