Looking at actual satisfaction scores post-FTX, we see more engaged people were more highly satisfied than less engaged people. In comparison, for current satisfaction, this is no longer the case or is only minimally so (setting aside the least engaged who remain less satisfied than the moderately to highly engaged). Every group’s satisfaction has decreased, with moderate to highly engaged EAs’ satisfaction declining to similar levels (implying a larger decrease among the more highly engaged).
The pattern is roughly similar, but less clear, looking only at changes within matched subjects (smaller sample size).
Looking at people’s recalled post-FTX satisfaction, there is no significant difference between the moderately to highly engaged (though they weakly lean in the opposite direction). So the recalled vs current comparison implies a slightly bigger positive gap for more highly engaged EAs (though we did not formally test this comparison).
For reasons for dissatisfaction, there are a few systematic differences across engagement levels:
More highly engaged respondents are more likely to mention Leadership
More highly engaged respondents were more likely to mention scandals
More highly engaged respondents were more likely to mention JEID at the lower importance levels (Important or Slightly important vs Very important), but it’s a less clear pattern at the Very important level
More highly engaged responents were more likely to mention Epistemics as Very important
The most highly engaged respondents were much more likely to mention Funding (though still less than the top factors)
Looking within the most highly engaged only, this that Leadership and Scandals are at the top, followed by Cause prioritization and JEID receiving similar levels of mentions.
Thanks!
For satisfaction, we see the following patterns.
Looking at actual satisfaction scores post-FTX, we see more engaged people were more highly satisfied than less engaged people. In comparison, for current satisfaction, this is no longer the case or is only minimally so (setting aside the least engaged who remain less satisfied than the moderately to highly engaged). Every group’s satisfaction has decreased, with moderate to highly engaged EAs’ satisfaction declining to similar levels (implying a larger decrease among the more highly engaged).
The pattern is roughly similar, but less clear, looking only at changes within matched subjects (smaller sample size).
Looking at people’s recalled post-FTX satisfaction, there is no significant difference between the moderately to highly engaged (though they weakly lean in the opposite direction). So the recalled vs current comparison implies a slightly bigger positive gap for more highly engaged EAs (though we did not formally test this comparison).
For reasons for dissatisfaction, there are a few systematic differences across engagement levels:
More highly engaged respondents are more likely to mention Leadership
More highly engaged respondents were more likely to mention scandals
More highly engaged respondents were more likely to mention JEID at the lower importance levels (Important or Slightly important vs Very important), but it’s a less clear pattern at the Very important level
More highly engaged responents were more likely to mention Epistemics as Very important
The most highly engaged respondents were much more likely to mention Funding (though still less than the top factors)
Looking within the most highly engaged only, this that Leadership and Scandals are at the top, followed by Cause prioritization and JEID receiving similar levels of mentions.