Do better schools have smaller graduation gaps?
This week we’ll continue our examination of opportunity gaps in the WA K-12 system by looking at school-level effects. We’ll start with a relatively simple question: Do “better” schools have smaller graduation gaps? I was rather surprised to find that the answer is “no”. In fact, for just about every group, graduation gaps are larger in better schools. If you don’t care about the details, you’re done! However, please note that there is an important caveat about focusing on “gaps” at the end of this post. For those of you interested in the details, read on.
As we learned in last week’s post, the low-income group has by far the biggest opportunity gap when it comes to graduation rates. To help understand school-quality effects, I’ve plotted low-income graduation rates versus non-low income graduation rates for individual schools below. It includes all schools in the state, since the patterns here seem to hold regardless of school type. A couple of things can quickly be seen from the chart:
a) Schools that have higher graduation rates for non-low income students also have higher graduation rates for low-income students. Obviously a lot of factors determine graduation rates, but we’ll just call the higher graduation rate schools “higher quality” for now. In other words, higher quality schools tend to have better results for all students.
b) There is a steady and persistent graduation rate gap between low-income and non-low income students, even for the the higher quality schools.
To illustrate b), I’ve also made a plot highlighting the graduation gap (the difference in graduation rates) for high quality schools, which I’ve defined as those with non-low income graduation rates over 0.7. I’ve also plotted a linear fit to the gap, which highlights that the average gap increases as school quality goes up.
We can can run the same analysis for different groups; I won’t bore you with all of the graphs, but they all generally show the same pattern: persistent, slightly increasing gaps as school quality increases. However, this raises an important question: how should we think about gaps relative to absolute graduation rates?
An important caveat on education gaps.
I was recently talking to someone about a particular program’s potential to increase education gaps. He suggested that if a program increased education gaps, it meant the program was harmful to disadvantaged groups. This is a pretty easy fallacy to fall into. Something can be good for disadvantaged students, while simultaneously increasing opportunity gaps. It simply means that the thing is good for disadvantaged students, but even better for “regular” students. The data above suggests that putting students into better schools is one of those things! It is important to remember that at the end of the day, we should generally care most about raising absolute, not relative achievement.
I think the fallacy is easiest to expose if you look at it it in reverse; i.e. making things worse. Would we want to put all of our students into worse schools, but ones that have smaller opportunity gaps? That answer is clear, but for some reason things seem a lot less obvious when we look at improvements. In this case, I find it helpful to remember that “a rising tide lifts all boats”. The analogy isn’t perfect, but it generally works. I think “gap analysis” is very useful as a diagnostic indicator of where improvements likely exist; but it shouldn’t obscure the main goal of improving absolute outcomes.
I recently read Kurt Vonnegut’s “Harrison Bergeron”, an extreme and silly exploration of focusing on relative gaps at the expense of absolutes.
Why would better schools have larger gaps?
Finally, I’ll offer a couple of hypotheses on why opportunity gaps may be larger at better schools, though clearly this topic deserves a deeper analysis.
The answer could be related to the fact that higher quality schools tend to have less low-income students; a topic we’ve explored in previously. Because they are a smaller fraction of the population, it could be that low-income students simply don’t get as much attention relative to when they make up more of the student body.
Alternatively (or in addition), the answer may have to do with students taking unequal advantage of resources. As school quality goes it up, it may become more difficult/complicated to take advantage of all the resources that the school offers. Disadvantaged kids may be less likely to make use of the full set of resources available to them. Thus the increasing gaps we see may be highlighting that we need to do a better job of making school resource easier to use…especially at higher quality schools.
In any case, this is a topic we’ll explore further in future posts!