Usability Testing: A Field Test Report
by Ting Ting Cheng
Home / Archives / Back Issues 1992 /
The following is a brief description of how I tested a user manual for a word processing program to be used by low-level and intermediate-level users.
First, I determined what I wanted to find out. I needed test subjects who would be representative of my intended audience. I wanted to know if the users of the manual could use it to learn to enter the program, then create, edit, print, and save a document I also wondered if I had too much or too little conceptual and/or procedural information, if the users could understand the instructions for a particular feature, and if the users could easily find the information they needed.
Usability testing can be done either in the lab or in the field. In either case, the subject performs tasks in a real-life context. For example, I planned to have the subjects type a letter, performing each of the basic procedures in any order they wished.
To test the instructions for a particular feature, I requested that the subjects use that feature, even when there was another way to perform the same function. For example, I had the subjects use the program's capability for printing while typing, an optional function that operates much like a typewriter.
The subjects were asked to verbalize their thoughts during the test. I recorded their comments and actions for later analysis.
I planned to interview the subjects to ensure I had not made false conclusions and to gather any additional information. Some of my questions were: What problems did you have using the manual? What did you read or skip? Was it easy for you to find the information you needed? Was there too little, too much, or adequate conceptual coverage? Procedural information?
After explaining the purpose of the test and the test activities, I warned them to expect no help from me in performing the tasks. I would only answer clarification questions.
I then stressed that the manual, not the user, was being tested. I tested and interviewed each subject separately so the viewpoints would not be biased.
The usability test was successful in pointing out weaknesses in the manual and in confirming its strengths. I discovered, for instance, that the symbol I used to alert the audience to a practice exercise -- a large exclamation point -- was frequently misinterpreted as a warning symbol.
In a comprehensive report on the usability test, I described: the test goals, test subjects, test method, test and interview results, and the manual revisions to be made based on these results.
From my experience a usability test gives concrete evidence of what works for your intended user. Why wait until the manual is printed and shipped out before you find out what's wrong with it? Why fix it later when you can fix it now? So, have you tested your manual lately?
Ting Ting Cheng is a Technical Communication graduate student at the University of Minnesota.