Some Random Studies I Found Interesting

Here are some random studies I found interesting. They’re unrelated to any argument I wanted to make or any more coherent piece I wanted to write. Most of them are from the Journal of Personality and Social Psychology, which is my favourite journal to browse. I find an interesting paper every time I read it. Enjoy!

Why is being nice bad for your bank account?

In discussions on the reasons behind the Gender Wage Gap, some people (read: Jordan Peterson and people who watch Jordan Peterson on YouTube) have claimed that women are higher in trait agreeableness than men and that being higher in trait agreeableness is linked to being less willing to negotiate your salary with your employer. Is this true? There’s actually not really much evidence to support it. Seemingly, researchers saw the negative correlation between trait agreeableness and financial health and came to their own conclusions about what might explain the link.

Matz and Gladstone (2020) provide some evidence that the reason that agreeable people are poorer is simply that they place less value on money, and not that they have more cooperative negotiating styles or anything like that. Surveying a few hundred people, they found that there is indeed a negative correlation between trait agreeableness and total amount of money saved. But they also found that there was no correlation between having a cooperative negotiation style and having more money saved. Rather, the relationship between agreeableness and low savings seemed to be mediated by how important someone considers money to be. People who say that they don’t really care much about money are both higher in trait agreeableness, and have less money. How highly people ranked the importance of money was measured using a simple 16 question survey.

They also did another study using representative data from the UK, finding an interaction effect, as can be seen above. What does this mean? The negative correlation between savings and agreeableness seems to be stronger among those with low incomes. The reason for this suggested by the authors is that if you’re poor, placing a low value on the importance of money seems more likely to be harmful, given you have more opportunities to squander your cash. You can decide whether you think this explanation is plausible or not.

The authors actually do a load of studies (in one paper) to show that the negative correlation between agreeableness and savings is mediated by placing a low value on money, rather than having a cooperative negotiation style. It’s only one paper, so take things with a pinch of salt, but it looks good. That being said, reverse causality seems like it could be a problem here (what if people with lower savings claim that they place a lower value on money?), and it’s all correlational evidence, so I wouldn’t bet the house that these results show a causal effect, but it’s fairly convincing for what it is. Here’s an excerpt from their conclusion:

The outlined results replicate across different samples, mea- sures, and methodologies; including nationally representative survey responses, digital records collected from customer bank accounts, online panels, and government-recorded insolvency rates across geographical areas in both the United States and United Kingdom. While each individual method has specific limitations (e.g., self-reports suffer from response biases, customer bank account records are not representative, aggregated geographical data limit inferences about individual behavior), the consistency of findings across samples provides strong evidence for the robustness of our effects. 

They also note that trait agreeableness is the only personality trait that is consistently negatively correlated with financial hardship across all their samples, which I found interesting. I kinda would’ve expected trait neuroticism to be negatively correlated with financial health, but seemingly not.

Notes on Moral Foundations Theory

You may have read Jonathan Haidt’s book The Righteous Mind, which goes into detail about moral foundation theory. If not, here’s a Ted Talk he did that gets to the gist of it. The basic idea is that there are five (or possibly six) different ways that most people decide whether something is right or wrong. The original five proposed foundations were harm, fairness, loyalty, authority, and purity. So, if someone is deciding whether an action is right or wrong, they might look at whether the action caused harm, whether it seemed to be unfair in some way, whether it was an act of disloyalty, and so on. 

The classic finding is that most people on the left are mainly just interested in the first two foundations. They think an act is wrong if it harms people or if it’s unfair, but they’re much less interested in whether someone isn’t respecting authority, or if they’re not being loyal (assuming that the disloyalty doesn’t also cause harm, which they are interested in). Conservatives, on the other hand, are interested in all five moral foundations. Or are they? Well, I think it’s more complicated than the original research suggests. 

Davis et al. (2016) seem to find that these results don’t really replicate in samples of black Americans, even though they’re still robust when the samples are primarily white. They carry out two studies, and in the first find that the correlation between conservatism and caring about authority and purity is much stronger in white Americans than in black Americans. In the second study, they seem to find no correlation between conservatism and either the purity or the authority foundation at all among black Americans, which is interesting. It sort of makes sense, I guess? The authors speculate that black people in the US are more religious than white people, and that being more religious is correlated with caring more about loyalty, authority, and purity (called the binding foundations). So, maybe the real connection is just between religiosity and the binding foundations, rather than conservatism and the foundations?

And some stuff from Political Science

I used to have an Anki deck that I used to memorise interesting political science studies. Here are a few results I think are interesting: 

When a child turns 18, their parents become more likely to vote (at least in Denmark, anyway). This study uses a nice Regression Discontinuity Design, which means you can take the results as a genuine causal finding. The authors show that having a child of voting age increases the chance of a parent voting by 2.8 percentage points. Again, this isn’t some weird correlational thing, it’s almost certainly a causal effect of a child turning 18.

Here’s an experiment published in AJPS that shows that investing in the stock markets makes people more likely to have believe that income inequality shouldn’t be reduced, and have more typically right-wing views on whether people deserve their wealth. The explanation here is likely to do with decreasing distrust of stock markets, and an adoption of the view that stock market returns are deserved. The treatment effect here is about 9% of the gap between Labour and Conservative voters, so pretty large. That’s kinda weird to me, I have my money in stocks and it seems extremely clear that any gains I make are just a result of me having money, but hey ho. 

Rose and Weßels look into the connection between referendums and populism in seventeen European countries, finding that the connection between being a populist and supporting direct democracy is probably a bit overstated. Weirdly, there is a negative correlation between thinking we should ignore the rights of minority groups and supporting more referendums.

There isn’t any correlation between being poorly educated or on a low income and supporting referendums. There seemingly is a small correlation between being on the far-right and supporting referendums, though. It’s also interesting that most people seem to be supportive of more referendums - 61% of people think that referendums are a good way to settle important political questions.