August 16, 2011

More mean than average

As we all know, mean people suck. But do they earn more?

A US study presented at a management conference today looked at measurements of agreeableness, and found that people (or, at least, men) who rated themselves as less agreeable, cooperative, and flexible earned more money.  This isn’t precisely about `mean’ people, but headline writers around the world spontaneously went for the four-letter word (or just copied each other).

The researchers looked at the relationship between income and self-rated personality questions in three longitudinal studies (NLSY, MIDUS, WLS) that had been carried out for other purposes and had made their data publicly available. In all three studies, people who rated higher on agreeableness had lower average income, and this wasn’t true (or was much weaker) for other psychological constructs.

As the researchers admit in their paper, they don’t know whether disagreeable people are more productive, or whether they are perceived as more productive by employers, or whether they are just better at extracting raises. Another possibility they don’t raise is that people who earn more money are just more likely to be honest about their lack of agreeableness, and those who earn less don’t like to admit to being non-nice.

In any case, the most interesting statistical issue is how the magnitude of the association gets reported.  TVNZ says that less-nice men “earned nearly 20% more a year than their workmates who described themselves more favourably.  That equates to more than $US9000 ($NZ10,800) a year more in pay. The Wall Street Journal was even more precise:  men who measured below average on agreeableness earned about 18% more—or $9,772 more annually in their sample—than nicer guys.

We can contrast this with the correlations between income and agreeableness in the paper, which were -0.11, -0.19, and -0.14 for the three studies. Squaring these to get the proportion of variation explained gives 1%, 3.5%, or 2%.   So where did the $9772 and 18% pay differential come from?   This is an estimate comparing people who are one standard deviation above the mean with people one standard deviation below the mean, averaged across the three studies. That is, $9772 is the predicted difference in income for two men who differ by two standard deviations in agreeableness, which is a big difference.  If you pick two men at random, the chance that they differ by more than two standard deviations is only about one in six.

There are also at least two reasons to be sceptical of how these results generalize.  Firstly, more-educated people rated themselves lower on agreeableness, and earned more, so part of what we are seeing could be an effect of education.  The researchers tried to adjust for this, but at best they had number of years of education, and in one study only had whether or not someone was a university graduate.   The second reason is that even if it’s true in the US, it may not generalize.  The difference in income with agreeableness was quite a bit smaller than the difference in income between men and women, and as we know, this difference varies enormously around the world and is smaller here than in the US. Standards of niceness also vary a lot from place to place — and two of the three studies were done in Wisconsin.

Thanks to Tony Cooper for the pointer to the reports and study.

avatar

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »