Wall Street Journal published an awesome editorial yesterday called The United States of Mind, which tracks the rankings of the Big 5 personality traits (openness, conscientiousness, extraversion, agreeableness, neuroticism) across each of the 50 states and Washington DC. The best part of this is the interactive page right here that lets you see the breakdown of the ranking for each state.
Well, I mean really, what else am I going to do with this information except analyze it? So, I’m sure whenever I get bored I’ll find a new metric I want to rank against these 5 and look at correlations and put them up here.
So, for starters today, we’re going with GDP per Capita, because what’s more important than money?
First though, a quick primer on what all of this means:
Correlation is how closely things are related and is measured on a scale from -1 to 1, with 0 being neutral (no effect). The closer a correlation is to 0, the less of an effect the two variables have on each other.
A correlation of 1 means that the two variables are directly related, so they go up exactly together. A variable always has a correlation of 1 with itself. A number like .8 means that two variables are very closely positively related. Height and weight have a very high correlation because tall people tend to be heavier.
A correlation of -1 means that two variables are inversely related, so that when one goes up the other goes down the same amount. Years of schooling and number of welfare checks picked up is probably a good example of a highly negative correlation (although I don’t have any information to back that up in front of me).
R-squared is simply the correlation coefficient squared. It basically tells you how much of an effect one variable has on the other. Note that the values won’t always add up to 100%, which means that there is some fraction of variance unexplained, or due to factors outside of the variables shown.
Finally, two major points. First, please don’t think that these factors are in any way the only ones affecting each other. There are a ton of variables that aren’t being accounted for here, and this should be taken as nothing more than an academic exercise. Secondly, and God knows this has caused people much smarter than you to make horrible decisions so pay attention here, CORRELATION DOES NOT IMPLY CAUSATION. There is no way to say that these numbers show that more openness leads to a higher GDP per capita or that a higher GDP leads to more openness. In fact, both of these may be due to some unknown variable, so please please please keep that in mind.
With all of that covered now, here’s the good stuff. If the images are blocked by a sidebar or something, click on them to open the image in a new window:
So, what can we see here? Well, there’s really only one pair I would consider significantly correlated and that’s the agreeableness/conscientiousness pair. Agreeableness, conscientiousness, and extraversion are the most strongly correlated variables on the whole, and they are all positively correlated with each other. The most surprising result to me is the negative correlation between openness and extraversion. It’s not that high, but the fact that it even exists doesn’t make much sense to me. Neuroticism is really interesting to me, as it seems to have little to no correlation with any of the other traits. In fact, excluding the GDP pairs, it has 3 of the 4 lowest correlations attached to it.
As for the 6th variable, openness is the only one of the big 5 traits positively correlated with GDP per capita. Both agreeableness and conscientiousness are about as strongly negatively correlated and openness is positively correlated, while extraversion and neuroticism are two of the most insignificant correlations on the entire chart.
So, there you go. There’s round 1.
