Pilot Study
On Friday we ran a first Pilot study to ensure that our systems were working as intended. For the most part it worked out well; we were to run a full test from start to finish without any major issues.
A key part of these studies is that the people that use the system should be anonymous strangers; they should not be known to us or each other. We're using Prolific to recruit and compensate people for taking part. Each study is about 30-45 minutes and involves a bit of instructions, a short and focused discussion, and a task that helps us understand how the discussion went from the participants' perspective. It's a combination of quantitative and qualitative data that we can use to compare variations of the system and understand which parts are having what kinds of effects.
There are just a couple of small tweaks to make to the experiment system, and then we'll be jumping into running a series of studies tomorrow that will give us some real data.