Crash Course: Usability testing II

If you haven’t done so yet, please visit my introductory piece to usability testing HERE. It may help make more sense as you read the following article.

GOALS

First, it is important to set goals for the test. Typically, these involve statements like the following:

  • I will single out parts of the site that are not intuitively navigable to the user
  • I will determine areas that hinder a user’s access to quick information

Write these down so that you can keep these in mind as you test the site.

MATERIALS (You can refer to this site for good examples of the different materials and forms you may need)

  • You’ll need five test subjects. You don’t really need many more than this to get functional results.
  • A working computer with internet access for all test subjects
  • Informed consent forms for all test subjects (PRINTED and SIGNED)
  • Pre-test (PRINTED)
  • List of assigned tasks (PRINTED)
  • Post-test questionnaire (PRINTED)
  • Post-test list of open-discussion interview questions

PROCESS

1. Determine tasks for test subjects to complete to target areas of supposed deficiencies.

  • There should be between 10 and 15 of these that are carefully selected to be of use to your experiment. In other words, designate these tasks to mimic what a user might actually go to the site to do. These can include things like: find a news story about dogs in Chapel Hill if it’s a news website, or “you have a comment you would like to make about a story you just read. Please write a comment on a story on the site.”

2. Have test subjects read and sign an informed consent papers

  • This explains what the users will be doing and for what purposes, and is ethically necessary to proceed with your test.

3. Instruct users to complete the pre-test questionnaire

  • Collects data about users for your purposes so you can include it in your conclusions after the experiment.

4. Allow the users to take some time to freely explore the site

  • During this exploration, there should be no interaction with the test-giver. Give them as long as they would like and record what they do with this time. After around 10 minutes (but usually they will be done before this) notate that they took this long and instruct them to move to the next step of the test.

5. Give users the tasks you outlined in the first step.

  • You may choose to read them the tasks or have them complete them from a list that they can read. Continue to record their actions and time them to see how long it takes to complete each task. This information will be useful to look back on to determine whether or not it is consistently taking users longer to complete a relatively easy task. Later, you may decide to redesign this part of the site accordingly.

6. Instruct the subject to fill out the post-test questions you’ve already created.

7. Involve the subject in a verbal discussion (you may want to electronically record this part) about the tasks they’ve completed.

  • Refer to the interview questions from the materials list
  • These questions should be open-ended and deal with how the user felt about completing the different tasks. Try and make this as conversational as possible so they feel at ease and honest.

6. Analyze the results of the testing

  • Decide which areas of the site need to be improved and which are working just fine. In this step, you may also wish to prioritize what needs to be fixed.

TIPS

  • Do not test people who know about the experiment or who are very familiar with the website you’re testing
  • Be consistent with testing methods between each subject so you don’t inadvertently skew the results
  • Be honest and detailed when taking notes on how a user completes each task as well as during their free observation period

Obviously this is a simplified outline of how to successfully conduct a usability test for your website or a freelance gig. However, it really isn’t rocket science and the more websites conduct these tests, the more improved our flow of information and communication will become.

Feel free to comment on anything I’ve missed as well as your own experiences with usability testing and what works the best! I hope that you’ve found this article educational if not just a good basic guide to usability testing.

About these ads