Read on Twitter
PNAS just published a paper titled "Experienced well-being rises with income, even above $75,000 per year". The author did not publish the raw data, but luckily, there's enough info to simulate it. I wrote a blog: https://lindeloev.net/new-pnas-paper-income-is-a-poor-way-to-improve-well-being/ and this thread, 1/7
The central figure shows an impressive linear relationship between log(income) and two measures of well-being. This is from a huge experience-sampling dataset from 33.391 participants. 2/7
Two things to note here. (1) the x-axis is logarithmic. There's a HUGE difference in income between the rightmost points. (2) the y-axis spans a TINY effect size. And TINY / HUGE = EVEN TINIER. 3/7
The z-score is just the number of standard deviations that an individual lies from the population. So we can simulate data (dots) from the standard normal distribution on these z-values. With a linear x-axis, we get a more representative plot: (see #rstats code in the blog) 4/7
A randomly sampled highest-income participant ($480.000) would have *lower* well-being than a randomly sampled lowest-income participant ($15.000) 25% and 33% of the time for the two outcome measures. Income explains 1.5% and 4% of the variance. 5/7
This is yet another reminder that *effect sizes matter*. Theory should point us towards the phenomena that *matter*. We cannot include all significant predictors in all our models. 6/7
In summary, I think this paper presents convincing evidence that household income over, say, $30.000 makes virtually no difference for well-being and life satisfaction. The Beatles foresaw this 57 years ago:
Whoa, the (heavily aggregated) public data doesn't even match the figure from the paper: https://twitter.com/kjhealy/status/1353857217905176577.
There's a great discussion going on in response to this and I'm learning. Especially about cases where small effect sizes *can* matter. E.g., this mini-thread by @hardsci, and the quoted thread by @rlucas11. https://twitter.com/hardsci/status/1354123109981282304
