See here for a more detailed synthesis of our assumption tests and experiment design.
Assumption Test 1: User Screen Time Customization
Pictures of testing:
Setting up Daily Assumption Study Survey
Setting up self-customizable screen time
Experiment Overview:
For our experiment, we divided our recruiter participants into two groups – one group where we enforced a screen time limit for them, and another group where we allowed the group to set their own. For each group, we gathered their screen time data (quantitative metric) and their feelings towards the app (qualitative) at the end of the week. This was our way of measuring “long-term impacts” given that this class is limited to 10 weeks.
Who we recruited:
We recruited users who matched one of our original user personas, Decreasing Desires Daisy – moderate users who consistently use TikTok throughout the day without any extreme spikes, with the goal of using it less frequently and intensively. We ended up recruiting 4 Stanford students who matched this TikTok consumption model.
Learning Card 1:
From this assumption test, the main takeaways we learned were that we want to help our users decrease their TikTok usage but in a way that doesn’t make them feel more stressed out about their screen time and take away from their time to relax by using TikTok. Our next steps that we decided on was to let users decide their screen time limits themselves, but stress the fact that they should choose a limit that actually challenges them to decrease their TikTok usage.
Assumption Test 2:
Experiment Overview:
For our experiment, we divided our recruiter participants into two groups, where the users in both groups input a predetermined amount of time to watch TikTok within one continuous sitting. One group texted a friend each time the alert for the end of their session was reached while the other (control) group did not. For each group, we collected screen time data (quantitative metric) and their feelings towards the alert with emphasis on how notifying another person impacted their behavior (qualitative) at the end of the week. This was our way of measuring “long-term impacts” given that this class is limited to 10 weeks.
Who we recruited:
Similar to our other experiment for this assumption, we recruited users who matched one of our original user personas, Decreasing Desires Daisy – moderate users who consistently use TikTok throughout the day without any extreme spikes, with the goal of using it less frequently and intensively. We ended up recruiting 3 Stanford students who matched this TikTok consumption model.
Artifacts:
We used the same Google Form as the above assumption test for gathering our user data and also conducted exit interviews for our participants, asking them to focus on how needing a friend’s approval impacted their TikTok usage.
Setting up Daily Assumption Study Survey
