https://www.usability.gov/how-to-and-tools/methods/running-usability-tests.html
In another article from Usability.gov, the procedures and techniques to conducting a proper usability test are outlines. The article immediately addresses two actions that must be conducted prior to your test, though. It mentions that the test should be completely planned and your test participants should be recruited and ready to meet you for the test. Once these two objectives are met, then you can begin the procedure.
The first thing that must be done is selecting a moderating technique. Though there are many styles for moderation, a popular technique is what’s called the “Concurrent Think Aloud” technique. Utilized heavily by Steve Krug in his books on usability, this style relies on participants sharing their thoughts out loud as the move through the usability test. This can be anything from “oh, I found it!” or “hmmm, now I wonder where I can find the contact page for this shoe store.” The point is, the participants share their real thought process as the move through the various tests, which is useful for the web designer trying to make improvements on the other side.
Other testing styles include “Retrospective Think Aloud,” which asks users to retrace their thought process post-test, “Concurrent Probing”, which involves the tester asking the participant questions anytime they mention something interesting, and “Retrospective Probing,” which involves the tester asking the participant questions once the testing has concluded. There are pros and cons to each one, but as the article mentions, you can’t really go wrong with any of the tactics. What is important, though, is being thorough with the way you record information and allow the participant to carefully run through each task.
Next, you’ll want to run a pilot test 1-2 days before the real test. This will involve testing equipment to ensure it isn’t faulty, practicing taking notes and asking questions, getting a sense of how effective your test structure is, and making any final adjustments.
The final section of the article discusses some of the best practices for a good usability test. Priority should be on making the test subjects feel comfortable, being thorough with both your questioning and tasks for participants to complete, and to take as many notes as you can.
Reading this article was useful. I’ve usability tested before, but this served as a reminder of what considerations I should make when developing a test. It never even occurred to me that there were different test styles, but it’s something I’ll consider modifying the next time I conduct a usability test.