I found it a little intimidating being the user in this scenario – I felt that there was a real pressure to get through the flow of the product quickly and smoothly. It was frustrating having someone takes notes even when you felt they weren’t warranted. In the case of the evaluator testing their own work, you had to remember that the notes were to help them improve the prototype, not judge your capabilities with their prototype. It was easier being the user when you were being tested on the site that no one had any affinity to (officeworks)
This was fairly straightforward, it was difficult to assess the level of detail to go into with this though e.g. does a small frown warrant as much weight on the score card than a bigger frown. I liked how assessing the facial expressions revealed very quantitative data, however, it was still fairly subjective data.
This was a little tricky because when the evaluator was testing their own prototype, you had to be sensitive to the fact that they had designed it so you didn’t want to be too harsh or too rude. It was a lot easier to think aloud on the generic site (officeworks). I found that once I got absorbed into the task, however, I needed to be reminded to think aloud because it is quite an unnatural thing to do
This was tricky when testing multiple participants because everyone approached “thinking aloud” differently and so the qualitative data between different participants varied greatly. It would have helped to provide the participants with more of a clear framework of the things to let me know as an evaluator perhaps – this comes as a trade off perhaps though, because you want data which is really accurate to what the user is thinking and not misconstrued by adapting the data for the kind of data I wanted to collect.