Report back on latest round of usability testing

At the end of June I ran another set of Usability tests with students.  The main goal was to compare the student experience six months after the first round,  now that we've addressed some of the issues highlighted back then.

The full report (and the videos, session reports and other associated materials) are on the JCU Library Intranet.

What did we learn?

• That the changes we made almost all led to improvements in the site's usability.
• On the issues where we had not yet made changes there was no significant change in performance with one exception (see The Exception below)
• Summon/One Search on the home page was used by all the testees (one used it to track down a libguide), as opposed to the previous catalogue search that was all but ignored in the first round.
• Terminology remains an issue, without successful information literacy training students don't make the mental connection between 'articles' and 'databases' and are far more likely to click on 'ejournals' to find articles on a subject. Reserve Online is still meaningless to them.
• Contacts still needs some work - task was performed with more success, but by paths not affected by our changes. Unlike the first round these testees went to the service to find the contact, unlike the first round who generally went to the generic contacts and tried to find those related to the service.
• Libguides seems to be accepted and utilised by students
• Remote access is still initiated by students as a separate process
• The form task (make a suggestion for purchase) once again displayed a 100% fail rate.
• That the DIY usability testing methodology has been validated as a method for identifying problems and measuring the effectiveness of our solutions.

The Exception
All the students knew how to change their passwords, almost certainly because the email changeover to Live@EDU at the start of the year required them to go to that page to test and/or change their passwords.

Now What?
Now I press on with developing a draft Information Architecture to resolve the issues identified that could not be fixed with page and link label edits. Using what we've learned from this testing, user surveys, feedback, usage stats and the Card Sorting Exercise.