Does Increased Public School Funding Lead to Improved Educational Outcomes?

No matter how hard we try, life never works out quite perfectly. And it’s in the complex and often chaotic world of K-12 education where this might be most obvious. I spent twenty years of my own life teaching high school and I can tell you it’s hard to get right.

One narrative heard often from inside the education industry blames all problems on lack of sufficient funding. On average, as it turns out, governments in developed nations already spend more than 22% of their gross domestic product (GDP) on their children’s education.

My gut tells me that the sudden introduction of unlimited funding might not live up to expectations. But no one should rely on my gut feelings. Instead, I’m going to use some publicly available data from the US and around the world to see if it can’t give us a glimpse of what works and what doesn’t.

Of course, I can’t guarantee that my interpretations of that data are objectively correct. Just because there seem to be striking relationships between data points doesn’t itself prove the truth of a given insight. Correlation, as you’ve no doubt heard, isn’t the same as causation. But an unproven insight that’s nevertheless compelling sure beats complete ignorance.

At any rate, as we’ll soon see, it turns out that the actual amount that’s spent matters less than how it’s spent. But I’m getting way ahead of myself.

Choosing metrics

Before looking for the right data to illuminate a topic, you need clear target metrics. In our case, what kind of data is likely to best represent student outcomes and government school funding levels? Once we’re clear about what we’re looking for we can head out to see if such data is available.

Measuring outcomes

What best describes “success” in the context of education?

I don’t think employment or income rates - even when focused on recent graduates - are useful: there are simply too many other macroeconomic and social factors at play. So attributing employment and income trends to any one particular educational funding variable would be nearly impossible.

What about college enrolment or graduation rates? It’s true that the products of a successful college-focused K-12 education will be better prepared for the challenges of higher education. But that would ignore the large non-college-bound segment of the student population, wouldn’t it? Perhaps a K-12 system that narrowly optimizes its curriculum for college success has failed to serve a major segment of its population. Which suggests that college success is a weak proxy for K-12 operations.

Using data measuring proficiency in science, technology, engineering, and mathematics (STEM) skills would similarly be too narrow. As important and profitable as STEM careers can be, they don’t come close to representing an educational system as a whole.

On the other hand, measuring literacy and numeracy (the ability to read and perform simple math) would be a meaningful metric but, considering how those basic skills are nearly universal across developed nations, they wouldn’t be helpful from the perspective of comparing outcomes.

So I guess that leaves the results of regular standardized testing of core educational skills. Sure, such tests are imperfect and there is always the risk that their existence can impact the very system they’re trying to measure. But they have both their advocates and unique charms. And they’re all we’ve got.

Specifically, we’ll be using two sets of testing metrics: