Epistemic status: sharing low-quality work just in case it’s useful for someone else. Personal, not professional.
A few years back, I was doing an MSc in Economics, and trying to take opportunities to do EA-ish research projects as I went.
There had been somediscussion about whether giving caused happiness, so I wrote a short paper that tried to look into that.
Basically, I did a few types of regression on a big US panel dataset, and it didn’t look like donations were correlated with life satisfaction. But maybe I messed something up.
(I then did some overly fancy regressions (instrumental variables) because signalling. This method didn’t really work and in any case I think I misinterpreted the variable in the dataset I was using—ignore that part of the paper.)
I don’t place much weight on this, and probably can’t remember the answers to technical questions that people might have. I’m just sharing because it seems like a good norm to default towards sharing things like this, and maybe it helps someone else to do a better version.
Some extremely rough research on giving and happiness
Epistemic status: sharing low-quality work just in case it’s useful for someone else. Personal, not professional.
A few years back, I was doing an MSc in Economics, and trying to take opportunities to do EA-ish research projects as I went.
There had been some discussion about whether giving caused happiness, so I wrote a short paper that tried to look into that.
Basically, I did a few types of regression on a big US panel dataset, and it didn’t look like donations were correlated with life satisfaction. But maybe I messed something up.
(I then did some overly fancy regressions (instrumental variables) because signalling. This method didn’t really work and in any case I think I misinterpreted the variable in the dataset I was using—ignore that part of the paper.)
I don’t place much weight on this, and probably can’t remember the answers to technical questions that people might have. I’m just sharing because it seems like a good norm to default towards sharing things like this, and maybe it helps someone else to do a better version.
Here’s the paper.