search

willschulz / bad-influence

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

Plan next data collection #22

Open JamesPHoughton opened 1 year ago

JamesPHoughton commented 1 year ago

Planning for next data collection should begin once we have completed the planned analyses of the initial data. This will allows us to:

  1. Identify and fix aspects of the procedure that may undermine data quality.
  2. Learn which types of content analysis work well out-of-the-box with a modest N (as opposed to methods we might want to build/fine-tune ourselves, and/or circle back to when we have a larger N)
  3. Plan a pre-registered (or at least well-focused) experiment that applies feasible analyses to (a subset of) our substantive research questions.
  4. Make a bigger-picture plan to iterate over this procedure to keep exploring more ambitious analyses and collecting more data in the long run.
willschulz commented 1 year ago

Sampling strategy will be part of this planning: I expect we'll be going to MTurk, and I also expect we'll want to create co-partisan groups