Is Ignorance Really Bliss?

A regular blog reader, Mitch Kosowski, sent along an interesting question: “Is ignorance truly bliss? Are people with lower intelligence happier than those with higher intelligence?”
Let’s start with a quick literature review. Here are the findings reported by Simpson, L. (2001):

Lisa Simpson: “As intelligence goes up, happiness goes down. See, I made a graph. I make lots of graphs.” [The Simpsons, episode 257]
Despite her formidable unhappiness, I don’t think Lisa is right on this one. My reasoning is simple: more intelligent people tend to earn higher incomes, and we know that people with higher incomes are more likely to be happy.
But that’s theory; let’s crunch some numbers.
The General Social Survey asks about happiness and also contains a simple vocabulary test, which we’ll use as a proxy for intelligence. While this is a pretty rough proxy, I’ll rely on the Rumsfeld defense, analyzing the data we’ve got, rather than the data we want. I simply divided people into the top, middle, and bottom thirds of the population, in terms of their vocabulary scores:

There’s also a small reasoning-based test:

Armed with these data, Lisa can make more graphs, and she’ll discover that those with stronger vocabularies or stronger analytic reasoning skills are more likely to be very happy, and less likely to be unhappy.
These differences were also statistically significant. By contrast, much of the existing literature finds no statistically significant relationship between individual happiness and intelligence. But the failure of small-scale studies to find statistically significant results likely reflects the fact that small-scale studies can’t verify much. (My analysis includes over 14,000 people; existing studies range from analyzing a couple of dozen to a couple of thousand people.)
Even so, these happiness differences look small. But I don’t immediately conclude that the happiness-intelligence link is weak; instead, these weak(ish) results may reflect a weak link between actual and measured intelligence. The coarseness of my intelligence measures means that there are likely some very intelligent and very happy people mis-categorized as moderately intelligent and very happy. If we could sort these people out, I think it’s pretty likely that we would find that there is an even stronger relationship between intelligence and happiness. (Hint for econ students: there’s an interesting paper waiting to be written on this.)
But this doesn’t answer the harder question: What creates a relationship between (measured) intelligence and (measured) happiness? Are those who are lucky enough to be born intelligent also lucky enough to be born happier? Do happy folks elicit greater attention from their teachers? Or does the sort of intelligence that is created by education also enable us to successfully pursue happiness? If it’s the latter, then perhaps these data point to yet another reason to invest in education.