7 ms·

I found that video incredibly unconvincing. It just feels like pseudoscience telling people what they want to hear, and the data was underwhelming (assuming it

I found that video incredibly unconvincing. It just feels like pseudoscience telling people what they want to hear, and the data was underwhelming (assuming it's not flawed to begin with). I think a more plausible explanation is that people just lie to themselves to believe whatever they want. These people are only "happier" until they're forced to confront a reality that contradicts their beliefs. In other words it's willful ignorance masquerading as synthesized happiness. There's also data backing up what I'm saying. There's a study that's regurgitated constantly saying people aren't happier after making a certain amount of money ($75k), but almost nobody bothered to actually read the study. I think this contradictory study is a lot more convincing. https://www.pnas.org/doi/10.1073/pnas.2016976118 https://www.pnas.org/doi/10.1073/pnas.2016976118

smokel 3y ago
Thank you for raising skepticism about old studies! The new one that you mention is slightly suspect as well though :) It is based on data gathered from trackyourhappiness.org. Why would someone install an app to track their happiness? I doubt that this is a representative group of people. Another interesting paper about this is "Income and emotional well-being: A conflict resolved", Kahneman and Mellers, 2022 https://www.pnas.org/doi/epdf/10.1073/pnas.2208661120 https://www.pnas.org/doi/epdf/10.1073/pnas.2208661120
aston 3y ago
You may be interested in this newer paper, reconciling the work of Killingsworth and Kahneman & Deaton. https://www.pnas.org/doi/10.1073/pnas.2208661120 https://www.pnas.org/doi/10.1073/pnas.2208661120
- gmac 3y ago
  I work in this area and I really dislike it when happiness is plotted against income only on a log scale, as is done here. The log scale effectively conceals the flattening-out of the relationship, which is the thing that shows you how progressive taxation + redistribution make the average person better off. Whether it's a straight line up to $75K followed by a flat line from there on, or a curved (log) relationship all the way up, seems to me to make rather little difference.
  - klft 3y ago
    Fig 2 of [1] shows high variance of happiness while income variation has only a small effect. There is roughly a 5 % (percent it is?) happiness delta for an income delta of about $ 500.000. However, the happiness delta between the different happiness levels is about 25 %. Would “Income and emotional well-being: It doesn’t really matter” be a better title for the study? Or do I miss something? [1] https://www.pnas.org/doi/10.1073/pnas.2208661120#fig02 https://www.pnas.org/doi/10.1073/pnas.2208661120#fig02
pawelmurias 3y ago
Measuring happiness by asking people to rate if from 1-10 seems pretty questionable.
- gmac 3y ago
  Do you have a better idea? (Happiness is a fully subjective thing, so subjective measures only, please).

smokel 3y ago

Thank you for raising skepticism about old studies! The new one that you mention is slightly suspect as well though :) It is based on data gathered from trackyourhappiness.org. Why would someone install an app to track their happiness? I doubt that this is a representative group of people. Another interesting paper about this is "Income and emotional well-being: A conflict resolved", Kahneman and Mellers, 2022 https://www.pnas.org/doi/epdf/10.1073/pnas.2208661120 https://www.pnas.org/doi/epdf/10.1073/pnas.2208661120

aston 3y ago

You may be interested in this newer paper, reconciling the work of Killingsworth and Kahneman & Deaton. https://www.pnas.org/doi/10.1073/pnas.2208661120 https://www.pnas.org/doi/10.1073/pnas.2208661120

gmac 3y ago
I work in this area and I really dislike it when happiness is plotted against income only on a log scale, as is done here. The log scale effectively conceals the flattening-out of the relationship, which is the thing that shows you how progressive taxation + redistribution make the average person better off. Whether it's a straight line up to $75K followed by a flat line from there on, or a curved (log) relationship all the way up, seems to me to make rather little difference.
- klft 3y ago
  Fig 2 of [1] shows high variance of happiness while income variation has only a small effect. There is roughly a 5 % (percent it is?) happiness delta for an income delta of about $ 500.000. However, the happiness delta between the different happiness levels is about 25 %. Would “Income and emotional well-being: It doesn’t really matter” be a better title for the study? Or do I miss something? [1] https://www.pnas.org/doi/10.1073/pnas.2208661120#fig02 https://www.pnas.org/doi/10.1073/pnas.2208661120#fig02

pawelmurias 3y ago

Measuring happiness by asking people to rate if from 1-10 seems pretty questionable.

gmac 3y ago
Do you have a better idea? (Happiness is a fully subjective thing, so subjective measures only, please).