Prototyping & User Testing in Learning DesignPosted: March 14, 2014
Prototyping and user testing are common practices in many design fields, yet when designing courses and curricula within higher education, developing prototypes and testing them with users seems to be a rarity. How might prototyping and user testing help us improve our designs for learning?
Prototypes help designers discover and address design challenges and constraints early in the design process through creating low-cost and often low-fidelity mock-ups. There is nothing like building a prototype to uncover the real challenges in a project. Prototypes can lead to solutions that could not have been imagined at the start of the design process, but through repeated efforts to improve a design, a beautiful solution is found.
Prototyping becomes even more helpful when we get our prototypes out into the world and test them with the people who will ultimately use what we are designing. Through user testing, we get incredible feedback on our design solutions. User testing gives us real answers to questions such as: Is it easy to use? Does it meet users needs? Did we make good assumptions about how users will approach our design? Through user testing, we get a view into the mind of our users so we can evolve and further develop our prototypes to better meet their needs.
When designing courses and curricula in higher education, we do have some practices that might be considered prototypes. Syllabi, curricular maps, and learning activity designs evolve through many drafts in the design process. Yet, are we treating these drafts as prototypes or are we refining a single solution? By developing multiple prototypes, each representing different approaches to a design challenge, designers take on a frame of divergent thinking-expanding and exploring many aspects and possible solutions to the challenge before moving to a convergent thinking mode to narrow and refine ideas. Yet, how often in learning design do we develop multiple concepts as possible design solutions? Often we skip exploring multiple approaches to a challenge and move immediately into narrowing and refining an early concept. Additionally, the process of developing multiple concepts requires a time commitment. Yet in learning design, do we give enough time to the prototyping process to allow designs to evolve and improve through the iterative building and refining of multiple prototypes?
While we may do some prototyping in course and curricular design work, I rarely see prototypes tested with students before a course or a curriculum is launched. There may be some good reasons for this, as many curricula take students several years to complete. User testing something so large prior to launch could be improbable or impossible for many institutions. Even at the course level user testing can seem impractical, trial running a semester long course for the purposes of user testing could be challenging for many schools.
Should we just skip user testing all together and make adjustments over time as we run the course with students year-over-year? This seems be to a standard practice in course design. We put together a syllabus, launch it, then gradually tweak and improve it each semester or year. Of course we should use student feedback to improve our designs over time, but don’t students in the early runs of a new course design deserve our best design efforts? How might user testing give us feedback to improve our designs prior to launch so that a course might be great on the first offering rather than having to wait a few years for it to be refined?
If doing a full semester or multi-year trial seems improbable, how might we do smaller, modular user tests? Could we develop user tests that allow us to test prototypes of particular learning activities or core resources? Could we find ways to test curricular structures and sequences with students without going through the whole sequence?
It seems we could learn a great deal in the design process and produce better designs more quickly by involving more prototyping and user testing in the process. So how do we do it? What have you seen that works well?