Social selection and social causation

Some of the most interesting puzzles for sociologists have to do with differences between groups of people, and two common explanations for social differences are “social selection” and “social causation.” They apply to a remarkably wide range of phenomenon, and they are kind of interesting to think about.

Here’s how they work. Suppose we have two types of people, those in social group “A” and those in social group “B”. Furthermore, we observe that the “A” and “B” people are different along some personal characteristic, say “X”.

How do we interpret this difference? If being in group “A” increases people’s “X”, then we are making a social causation argument. That is, we believe that being in that social group causes people to be different.

It could be, however, that being high on characteristic “X” makes people join group “A”. If so, we’re making a social selection argument because being the type of person “X” causes people to join a different group.

As such, the correlation between group “A” and characteristic “X” can be explained by both social selection and social causation, and what’s really interesting is to try to figure out which, if not both, are operating. To do so, you try to figure out which is the most plausible story and find any evidence that you can.

Here are some examples.

I know several people who graduated from Harvard, and they are some of the smartest people that I know. Freaking genius comes to mind. (Just for fun, I should say that one of them is a dunce just to keep my Harvard friends wondering if I mean them, but it wouldn’t be true. They’re all bright.) Why are they so smart?

It could be that Harvard provides a superior education. You have brilliant professors, world class facilities, and a rich legacy. If sitting in Harvard classrooms makes you really smart, this would be a social causation argument. Or, it could be that Harvard attracts the smartest students. The very brightest of high school students get their choice of schools to attend, and if they favor Harvard (and other top schools) then Harvard will produce very smart graduates, even if a Harvard education isn’t better than other schools.

Hm-m-m-m, how can we figure out which it is? If I were to guess, I would probably lean more to social selection. We know for a fact that Harvard attracts the best students. Ninety-five percent come from the top 5% of their high school classes, and they have SAT scores well north of 2000. Wow! (Digression. The best line I’ve ever heard about SAT scores—Jennifer Lopez was asked what she got on her SATs, and she replied “nail polish.”)

Harvard professors are brilliant, but they are selected for their ability as researchers, not teachers. Just because a professor creates new theories to revolutionize their field doesn’t mean they are particularly good at explaining the basics to students. They are also rewarded with salary and promotion for being star researchers, not for high quality teaching. I’m not saying they are bad teachers, but I can’t see how they would be much better teachers than those found in other four-year universities. If there is social causation, I would imagine it has to do with being around other smart students. Studying with, competing against, and talking to smart people will make you smarter, but this would be true with a group of people standing in a farmer’s field, it’s not a Harvard thing per se.

Here’s another social puzzle. There is a significant correlation between mental illness and lower social class; poor people are more likely to exhibit mental illness than rich people, including depression. How do we explain this?

A social causation argument would say that being poor increases mental illness. Not having enough money creates stress in people’s lives, and this stress can result in depression and other disorders. The poor are also less able to afford medication, counseling, and other ways of dealing with mental illness such that their conditions are more likely to get worse than those of wealthy people.

A social selection argument would portray the mentally ill as less able to get ahead in society. If you’re depressed, it may be hard to go to work regularly or to put a lot of energy into your career. People who have observable symptoms of mental illness might be less likely to interview well for a job or be promoted once they have a job. (An interesting question: Is this a form of discrimination?)

Which do I think it is? Well, I’m cheating here because I participated in a study that looked at this very question, and we found that both social selection and social causation mechanisms were in effect.

I’ve discussed two examples, but there are lots of other social phenomena that lend themselves to both social causation and selection arguments. For example:

Religious people live longer than non-religious people. Does religion change people’s life expectancy, or does it attract people who would live long already?

Criminals have more criminal friends than do non-criminals. Does having criminal friends make you into a criminal, or do criminals attract other criminals as friends?

Capitalist countries are wealthier than those with other types of economies. Does capitalism make a country wealthy, or do wealthy countries gravitate toward capitalism?

Interesting stuff, no? If you keep these two mechanisms in mind, you’ll be surprised by how many things you can explain by using them.

Originally posted in everydaysociologyblog.com