Besides the selection bias mentioned in the post and other comments, I’m also unsure about how much to update on people’s assessment of the mental health effects of their actions. I expect most people overestimate the positive mental health effects of any activity they choose to do.
As an extreme example, I expect that most smokers think that smoking is neutral or positive for their mental health, despite probably being net negative.[1]
Thanks for those analyses Lorenzo! I was aware of the claims that people don’t tend to be good at knowing what makes them happy or unhappy (thanks to an old 80K article!), but it didn’t occur to me that this could have influenced this data.
With this in mind it seems fairly possible that the real distribution should be more negative. Despite that, I still feel that the data feels like a positive update for me. a) There are just quite a few people reporting that being in EA is good for their mental health—and I don’t want to doubt that too much (but maybe that is hopeful thinking on my part) b) Also, before the survey results came out I expected that a larger number would report negative impacts as I had heard many people in the community reporting to me that EA had negatively affected their mental health (which is expected given my role), and I’d also heard some community builders tell me that they felt that EA was often bad for members mental health (which worried me a lot more).
It would be interesting to ask people in comparable groups, the same question of how involvement changed their mental health—if the distributions were a lot more positive or negative than EAs then that would be interesting. Maybe environmentalism or animal advocacy would be a reasonable comparison group as involvement might also give a similarly increased sense of awareness and obligation.
Besides the selection bias mentioned in the post and other comments, I’m also unsure about how much to update on people’s assessment of the mental health effects of their actions. I expect most people overestimate the positive mental health effects of any activity they choose to do.
As an extreme example, I expect that most smokers think that smoking is neutral or positive for their mental health, despite probably being net negative.[1]
I don’t know what the research says about this, but LLMs seem to agree:
ChatGPT Deep Research
Gemini Deep Research
Claude Research
Smoking is an extreme comparison, but the same happens with e.g. veganism
ChatGPT
Claude
Gemini didn’t only consider perceived effects, so it’s less relevant
Thanks for those analyses Lorenzo! I was aware of the claims that people don’t tend to be good at knowing what makes them happy or unhappy (thanks to an old 80K article!), but it didn’t occur to me that this could have influenced this data.
With this in mind it seems fairly possible that the real distribution should be more negative. Despite that, I still feel that the data feels like a positive update for me.
a) There are just quite a few people reporting that being in EA is good for their mental health—and I don’t want to doubt that too much (but maybe that is hopeful thinking on my part)
b) Also, before the survey results came out I expected that a larger number would report negative impacts as I had heard many people in the community reporting to me that EA had negatively affected their mental health (which is expected given my role), and I’d also heard some community builders tell me that they felt that EA was often bad for members mental health (which worried me a lot more).
It would be interesting to ask people in comparable groups, the same question of how involvement changed their mental health—if the distributions were a lot more positive or negative than EAs then that would be interesting. Maybe environmentalism or animal advocacy would be a reasonable comparison group as involvement might also give a similarly increased sense of awareness and obligation.