search

willschulz / bad-influence

Video discussion experiments to counter pathological influence in political discussions
0 stars 0 forks source link

Emphasize that participants can leave if they feel uncomfortable in the discussion #18

Open JamesPHoughton opened 1 year ago

JamesPHoughton commented 1 year ago

We want to make sure that the experiment is both safe for participants and encourages participation even when conversations get difficult. This is a bit of a difficult balance to strike, and we need to be careful about how we communicate our expectation to participants.

Right now, we give participants the options to mute their video and audio, or to leave. Do we want other options?

willschulz commented 1 year ago

I see two reasons to deviate from our current procedure:

  1. If we have reason to believe subjects are experiencing harm because they think they have to stay in the conversation.
  2. If the current procedure undermines external validity (ie if people would exit a real-world conversation, but remain in our discussions due to financial pressure.

We can assess (1) by reviewing some videos and doing a gut-check. We could also add relevant questions to the post-discussion survey, like "Was there any point when you wanted to leave the conversation?" and "If so, why did you stay?" (open-ended). This could help us assess both 1 an 2.