Methodology

Chapter 6 - Teaching Technology for Community
Managing IT in Government, Business & Communities
by Gerry Gingrich (ed)
Idea Group Publishing 2003

	Brought to you by Team-Fly

In building their projects, students learn and use the process of task-centered user interface design (TCD, 1994). In this process, the users' statements of their own needs and capabilities drive the product design. While developed for user interface design, task-centered design applies to the creation of any product where usability is the key concern.

Task-centered design follows these steps:

• Figure out who will use the system to do what

• Plagiarize!

• Rough out a design

• Think about it

• Prototype

• Test with users

• Iterate

  • Build it

  • Track it

  • Change it

Students begin by interviewing prospective users of their product, learning about their backgrounds, and about what they want the product to do. From the interviews, the students formulate a list of tasks. A task is a statement of something that needs to be accomplished that does not include information about implementation. A sample task for the events scheduling calendar might be "Mary, the events coordinator for XYZ organization, wants to change the starting time for an event she posted on the calendar last week." The collection of tasks defines the functionality of the product.

Working from the list of tasks, the students research commercial products and Web sites, looking for existing materials that relate to their projects. From those products, students liberally borrow features and functions (within the constraints of copyrights and patents.) The value of this "plagiarism" step is two-fold: borrowing reduces development time and borrowed features are often already familiar to users. The students then sketch out a detailed design on paper (a "low-fidelity" prototype) and proceed from the sketches to a rough computer implementation (a "high-fidelity" prototype). See, for example, Rudd, Stern, and Isensee (1996) for more information about prototyping.

In practice, evaluation of paper sketches proves too difficult for the students, so students instead "think about" or evaluate the high-fidelity prototypes. Project groups first judge their work using heuristic evaluation (Nielsen, 2002a). Each group member examines the prototype individually, looking for usability problems. The students are guided by Jakob Nielsen's ten usability heuristics, (Nielsen, 2002b) which are crafted to cover most known usability problems. The group members compare the results of their evaluations and determine how to correct the prototype in response to their findings. Each project group then carries out a cognitive walkthrough (Rieman, Franzke, and Redmiles, 1995) of the prototype. The students begin with a task statement and the list of steps needed to accomplish that task using the prototype. They walk through the list of steps, keeping in mind the interests and abilities of the users, and trying to tell a believable story about what the user might think and do at each step. Again, the students correct their prototypes to fix awkward or non-intuitive features. The two different types of evaluation help the students to identify different types of problems with their prototypes.

Carrying out the evaluations is typically difficult for the students as they, as undergraduates, are seldom asked to evaluate their own work. As a result, not all of the students appreciate the importance of the evaluations, and many are not patient enough to do a thorough job. They often believe that they have made only the best design choices and that there is no room for improvement. Their opinions change markedly, however, with the final round of thinking aloud (TA) evaluations (Lewis and Rieman, 1982). In contrast to the other two evaluation types, TA evaluations are done with users. In TA evaluations, the students observe a user trying to accomplish a task with the prototype. The user is asked to speak aloud all of her or his thoughts about the prototype during the process. Almost without exception, most of the user's thoughts and decisions (and stumblings and errors) catch the students completely by surprise. More than one student group has redesigned its prototype completely from scratch following the first round of TA evaluations, but responding to the users' actions and opinions typically gives the students a good sense of accomplishment.

The steps of high-fidelity prototyping, evaluating, and correcting the projects correspond to the last, iterative phase of task-centered design. Students repeat these steps until they and the users are satisfied with the results. Typically, the first round of TA evaluations has a large impact on the projects, while the second and third rounds deliver refinements.

	Brought to you by Team-Fly