Your Kindergarten Teacher and Your Salary

Kevin Drum has a post from a couple days ago discussing the correlation between early childhood education and professional success. The interesting part of the correlation is summed up by Raj Chetty, the Harvard Economics professor who co-authored the recent study, “How Does Your Kindergarten Classroom Affect Your Earnings?” Chetty says:

One thing I found interesting about the effect of test scores as relates to earnings is that it seems like some of the gains you find in early childhood, that might not show up on later test scores, later emerge when you’re looking at earnings data.

Yes, and that’s in fact, I think, the most striking finding…[Test scores] fade out over time. So kids who had better teachers and were in smaller classes in kindergarten aren’t doing all that better, really, on tests in middle school and high school. But what’s surprising is that those effects reemerge in adulthood. And I can talk about why we think that is.

Chetty goes on to give an explanation of the findings:

One explanation for this fadeout and then reemergence of the impact of kindergarten is through non-cognitive channels. […] For a limited subset of the students we have measures of non-cognitive ability in eighth grade. So what that means is measures like, they ask teachers to evaluate whether the students are being disruptive in class, whether the students are putting in a lot of effort, whether they’re motivated and so on. Now, we find persistent effects of your kindergarten class on these non-cognitive measures. There’s no fadeout, or very little fadeout on the non-cognitive stuff.

So one potential explanation of all of the findings together is, a good kindergarten teacher teaches you the material that you’re tested on in kindergarten, and so you do well on kindergarten tests. That same good teacher also imparts non-cognitive skills, like they teach you how to be a disciplined learner, how to put in a lot of effort, how to be patient….It’s quite intuitive that these non-cognitive skills matter when you’re an adult. It helps to get a good job and to do well in general if you’re a disciplined person, if you’re perseverant and so on.

Kevin Drum gives his reaction:

Bottom line: school matters, and the way it matters doesn’t get picked up entirely via standardized testing. In modern society, there are lots of behavioral traits that are just as important as IQ and subject matter knowledge. But we only test for subject matter knowledge, and so it gains an outsize importance.

In principle I agree with Drum that “school matters,” and that behavioral traits are undervalued in our society, but I think there’s a lot more to these findings than meets the eye. First, as with all studies that end up positing a correlation between two conditions, or two phenomena, there is a question as to which ways the causal arrows point. It’s one thing to say that there is a correlation between a nurturing Kindergarten experience and a high salary. It’s quite another to say that one causes the other.

And there are reasons why certain kids end up with better teachers and smaller classrooms in Kindergarten. For one, I think it’s true that wealthy kids are far more likely to be placed into nurturing early childhood education (ECE) environments than their poorer peers. They may not be as smart as their peers, hence the lower test scores in middle and high school, but with big parental investments later on in college and grad school, they are more than able to make up the lost ground intellectually and thus more likely to end up in higher-paying jobs. I don’t have the data to support this, but I think it’s a plausible explanation of Chetty’s data. And if so, it raises two points. First, high salary does not strictly imply intelligence or high academic performance (big surprise). Second, it might not be your Kindergarten classroom that is responsible for your high salary, but your parents’ socioeconomic status (SES) that is responsible for both.

That brings into question Chetty’s “non-cognitive skills” explanation. I agree that things like discipline, work ethic, speaking and listening skills, patience, obedience and so on, are critical to success in school, work, personal life, and beyond. I would also agree, given my own experience, that subject matter knowledge and quantitative performance gain an outsize importance during the middle and high school years (partially because of the trend towards fairer college admissions). But would Chetty really suggest that we should be more active about cultivating non-cognitive skills during middle and high school just so that more people can make more money later in life? That seems unsupported by the data, not to mention a wacky moral statement. If we want to focus more on developing non-cognitive skills in kids, it should be with an eye towards molding good people that are not merely effective professionals.

In any event, to say that there are “persistent effects of your kindergarten class on these non-cognitive measures,” is to mistake a correlation for a cause. The key question is why certain kids end up in the type of ECE environment that correlates with high salary. And if you delve into that question, it becomes unclear that your Kindergarten teacher has any effect on your salary, much less a big one.

Political Participation and the Internet

John Sides, via Ezra Klein’s blog, discusses a new study that shows that the internet does very little to balance out stratified political participation:

The graph above shows the amount of political activity for different levels of SES. Online political activity is as stratified by socioeconomic status as is off-line activity. The line for “offline act” ascends about as steeply as the line for “online act.” And this is not simply a function of Internet access – i.e., the “digital divide.” The line for online acts among Web users ascends almost as steeply. A similar finding emerges when the focus is donations to campaigns. Those donating online are doing so in smaller amounts, but these small donors are no less affluent than small donors giving offline.

If you think about this data purely in terms of the growing accessibility of the internet, I would admit that it is unnerving. But if you think about it in terms of the monetary cost of time, it’s not nearly as surprising. Signing a petition online may take less time than doing so at a local community center, but it still takes some time, and the fact is that people with lower SES do not have as much free time to engage in such activities as more affluent people. What this points out is something that has been true since ancient Greece: routine political participation is a leisure activity. For those members of society who don’t have – and can’t afford – much leisure time, it is difficult to sustain a high level of political participation, even with the increased accessibility and speed of the internet.

There is also a point to be made about access to technology. Low-SES people tend to have less up-to-date computer equipment and less knowledge of how to use the internet. So on top of having less time, low-SES citizens have fewer means to participate. Hypothetically, if the government were to extend free high-speed internet to all voting-age citizens, I think this chart would balance out somewhat. However, it wouldn’t balance out completely, and that brings us to the third and most important point.

Online political activity has to be preceded and complemented by offline political activity. Political participation is not a purely solitary activity like using Facebook or Twitter. Even in the case of signing a petition, which can be done privately online, there is no large chunk of the population that is signing petitions on their laptops and not engaging in politics in any other way. The reason is that politics is first and foremost a social activity. In order to care about political outcomes and develop personal opinions, we have to have face-to-face conversations with others (at least some of the time). The in-person component is essential to caring about politics, and unfortunately, it is not something that the internet can really facilitate. When you look at it this way, the internet is really just a new platform for participation, not a motivator of political participation. Thus, the people who are using the internet for political participation the most are the ones who already care about politics and have the time to participate by other means.

The problem comes full circle once you realize that engaging in politics in person requires an investment in time that low-SES people largely can’t afford.

What’s Wrong with “Kids” These Days?

From the NYT, a thought-provoking, though problematic, piece about directionless twenty-somethings:

It’s happening all over, in all sorts of families, not just young people moving back home but also young people taking longer to reach adulthood overall. It’s a development that predates the current economic doldrums, and no one knows yet what the impact will be – on the prospects of the young men and women; on the parents on whom so many of them depend; on society, built on the expectation of an orderly progression in which kids finish school, grow up, start careers, make a family and eventually retire to live on pensions supported by the next crop of kids who finish school, grow up, start careers, make a family and on and on. The traditional cycle seems to have gone off course, as young people remain untethered to romantic partners or to permanent homes, going back to school for lack of better options, traveling, avoiding commitments, competing ferociously for unpaid internships or temporary (and often grueling) Teach for America jobs, forestalling the beginning of adult life…

And the data to make your teeth chatter:

We’re in the thick of what one sociologist calls “the changing timetable for adulthood.” Sociologists traditionally define the “transition to adulthood” as marked by five milestones: completing school, leaving home, becoming financially independent, marrying and having a child. In 1960, 77 percent of women and 65 percent of men had, by the time they reached 30, passed all five milestones. Among 30-year-olds in 2000, according to data from the United States Census Bureau, fewer than half of the women and one-third of the men had done so. A Canadian study reported that a typical 30-year-old in 2001 had completed the same number of milestones as a 25-year-old in the early ’70s.

It’s pretty difficult to look past the nostalgia with which this piece is written: “The traditional cycle seems to have gone off course,” “young people remain untethered to romantic partners…” With all the changes that occurred in our society between 1960 and 2000, few of which are described here, is it really surprising that people take longer to settle into domesticity than they used to? Moreover, is it a bad thing? It’s one thing to make an emotional appeal to the past, but it’s another to disguise an emotional appeal as conscientious sociology. Presenting statistics out of context, and assigning rigid meanings to such malleable notions as “adulthood” and “growing up,” is not a good way to start an argument that attempts to describe a real social problem.

I will say this: if you look at the past hundred (or two hundred, or thousand) years, it’s pretty clear that the duration of childhood has been steadily growing, especially for the middle class. That is, the amount of time it takes for kids to become adults has been getting longer and longer, and not just since 1960. But this is not necessarily a cause for concern, and it is at least partially explained by a steadily increasing standard of living for the middle class, higher college enrollment rates, the feminist movement, contraception, and many other things that are are generally agreed to be good for society. In the not-so-long-ago days when more people went straight from high school to career-type jobs, it was easier to start a family and become a self-sufficient adult earlier. But now, more people have more time to spend on themselves – educating themselves, traveling, exploring different fields, living in different places, having different partners. Of course these things delay the onset of adulthood, but maybe they also allow for more thorough self-development and make for more responsible – and more experienced – adults in the long run.

I think the important thing to remember is that “adulthood” really can’t be talked about as if it’s an objective term across different social contexts (generations, cultures, etc.). Sociology should be about understanding how each generation defines and reaches adulthood, and what that means for the progress of society over the course of several generations. To suppose that our society has a serious problem on its hands simply because the new generation is taking longer to reach maturity benchmarks x, y, and z than the one before it, is, I think, to pervert sociology into an unreflective game of statistics.