Poverty and Educational Achievement: the importance of having enough

Everyone knows that, on average, poor kids do worse in school and on reading tests. Poverty is highly correlated with school success, and in fact despite the perennial search for unicorn schools that can magically bring poor kids up to the level of rich kids, I have never seen any evidence of a school that can bring a non-cherry-picked population of poor students to the level of achievement of many schools full of non-poor students. I've written about this here and here. Today I want to highlight a couple of studies that point to why poverty has such powerful effects. But first, I want to highlight a few common but misguided responses to the fact that poverty plays a huge role in educational achievement. 

The first response is to write off the difference as simply genetic, and therefore impossible to change. Beyond being probably just wrong, this response is not socially acceptable, so you don't see it said explicitly very often, but I do think it's an unspoken assumption that underlies some of our inaction.

Another response is to blame the individual students or their families. If they aren't rich, it's their fault, because they didn't follow Benjamin Franklin's precepts: work hard, be abstemious, and save money. Like the genetic response, this is a mostly unspoken excuse for inaction.

A third response is to blame teachers, or teachers' unions, or schools. This is absurd, since there is literally no school that actually manages to overcome poverty and educate non-cherry-picked students up to the level of low-poverty districts, but its absurdity doesn't prevent its being a favorite idea of education reformers everywhere. They get to cloak their elitism and greed in the guise of social justice.

A fourth response, which is in some ways the most interesting, is to suggest that what sets the rich kids apart is that they are given extra help and cultural "enrichment" by their families. I hear this one all the time from well-meaning colleagues. Rich kids get trips to Europe, and to museums, while poor kids go to the broken down playground down the street. Rich kids go to artsy summer camps in Maine, while poor kids go to the Y. Rich kids get tutoring, while poor kids sit in front of the TV. There may be something to this, but it makes it seem like the difference between poor kids and rich kids is that the rich kids get something extra, and I don't think that's true. 

It's not that rich kids are getting something extra, it's that poor kids aren't getting enough.

When I was a kid, my family did not go to Europe. I didn't attend expensive summer camps. My mom was on food stamps for a while. My parents worried about money. They ended up doing fine, but they were always incredibly frugal. I never had a tutor (though to be fair I did have highly educated parents who were around a lot). What I did have was about a million books, and lost of time to read them, and not many worries about my material comfort. I was pretty sure I was going to get enough to eat.

Two recent studies show the importance of getting enough:

1) Study 1: Kids read better when they're fed better

SNAP is the government program that gives low-income families a debit card with which they can buy food (people still refer to the program as "food stamps"). This is a great program, but it doesn't provide all that much money (less than $150 per person per month). The debit card is credited once a month, and families often find themselves running out of money as the month goes on.

In North Carolina, the money is credited to different families at different times, depending on the last couple of digits of the recipient's social security number. This allows for cool natural experiments like the one in the paper I read. A study by a Duke professor compares students' standardized test scores to when the students' families received their SNAP benefits. It turns out that scores are at their worst just before and just after the benefits are credited, and the scores are best about two to three weeks after the money arrives.

In the words of the study authors:

Student reading test scores appear to peak in the period from the 15th to 19th day post-SNAP receipt, and student math test scores appear to peak in the period from the 20th to 24th day post-SNAP receipt.

This interesting result suggests that nutrition matters to testing--and, by extension, very likely matters to learning as well. It doesn't matter a huge amount (something like 3% of a standard deviation), but it is easy to imagine a relatively small effect snowballing over time and creating a reverse Matthew effect, since if you fall behind by even 0.5% every year, by the time you're a senior in High School you will be behind by 6%, and it's likely that even at peak performance, kids whose families receive SNAP benefits will be less well-fed and more stressed in general.

2) Study 2: The stress of poverty takes up mental bandwidth

This study, by a few economists, two of whom wrote a book about this effect, is a bit further from education, but has larger effect sizes. These scholars did experiments designed to study how the stress of worrying about money affects cognitive performance. The results were more dramatic than I would have expected.

In one set of experiments, the authors asked shoppers at a mall in New Jersey to consider how they would handle a financial issue--for example, “Your car is having some trouble and requires $X to be fixed. You can pay in full, take a loan, or take a chance and forego the service at the moment... How would you go about making this decision?”)--and then, while they were considering the financial issue, to perform some basic cognitive tasks. If the financial issue was relatively easy (e.g. the car only required $150 to fix), then rich and poor people performed equally well on the cognitive tasks. If the issue was more difficult (e.g. the car required $1500 to fix), then rich people scored much, much better on the cognitive tasks, probably because the poor people were worrying about how they would come up with the $1500.

In another experiment discussed in the same paper, the authors gave cognitive tasks to small-holding sugarcane farmers in India before and after the harvest. This is a naturally randomized experiment, since different farmers harvest their sugarcane at very different times, over a several month period, according to when the sugarcane mills, which have a limited capacity, can process their cane.  The time before the harvest is not a time of poor nutrition, but it is a time of financial pressures (for instance, the farmers pawn items at a much higher rate). In this time of financial pressure, the farmers were much, much worse at cognitive tasks than they were during the relatively flush period after they harvested and sold their crop.

In these experiments, poor people who were thinking about money pressures performed far worse on cognitive tests than when they weren't thinking about money, and worse than people who had enough money that they didn't have to think about it.

Fortunately, these studies point to a relatively simple solution: reduce poverty!

It seems obvious to me that the solution to the poverty problem is not better education; rather, the solution to the education problem is less poverty. Just as we know, from studying public health, that the most important factors in a population's physical health and longevity are not the quality of the hospitals and doctors, so, in education, the most important factors in academic performance are not the quality of the schools and teachers. Just as my cousin wants to be an excellent doctor, so I want to be an excellent teacher. But as a society, we need to pay more attention to educational public health, and restructure our society so that, as in other rich countries, poor children are less poor. As a bonus, these policies would make us healthier, too!