Blog 4
Before we started testing the design, we put together some ideas for our user test script. This was useful as it got us discussing what we might say during our user tests and hence got us in the mindset. We decided that the script would involve two separate tasks with a series of subtasks which the users would be asked to complete. We also created scenarios to help get the user in the mindset of searching for a job.
We chose to recruit people that we knew who were over the age of 25 as we felt this fit the target group we had designed for. Before testing it was important that we sent the users consent forms to fill out.
Using the same script, we organised tests individually and recorded the sessions. Looking back, it may have been better to have solidified the way we were going to do our user tests as we each approached things slightly differently. I believe if we had done a practice user test with all of us present it could have brought up some issues which we could have discussed before the real thing.
We regrouped after the first round of tests and learned from our mistakes for the second round of testing. We took notes on our Figma file having discussed insights into flaws with the workings of the prototype as well as issues with the learnability of the app for our users. We assembled again on Figma and began to redesign the app again.
We decided that for the next round we would test in pairs with one person taking notes while the other interacted with the user. We then revised our script and made improvements based off of any issues we ran into previously and began another round of user tests.
During testing we asked our users to think aloud. This gave us some interesting insights into how the user felt about the app as they navigated through it (below). We also observed our users during the test as well as afterwards having recorded each test. This also gave us some helpful information which informed some redesign decisions.
As well as these qualitative results we also made sure to get quantitative data by creating a SUS questionnaire. We sent this questionnaire to each user following the test. Something which we could have done better here was to make sure that the user received and answered this questionnaire as quickly as possible. Some users didn’t do this, which may have altered the results as it could be said that they might not remember their experience as well. This questionnaire helped us to measure the satisfaction that each user got from the app and the results showed that we were on the right track. However, we only had a small sample size of people who completed the SUS questionnaire and it would have been better if we had gotten more responses. It has been said (Brooke, 2013, p. 5) that you need between 8 and 12 users in order for it to produce valid results.
References
Brooke, J. (2013). SUS: A Retrospective. The Journal Of Usability Studies, 8(2), 5–6. http://uxpajournal.org/wp-content/uploads/pdf/JUS_Brooke_February_2013.pdf
