Is Education Good for You?

Why hasn't anyone seriously tested education policy alternatives by analyzing moral outcomes

Back when I was a teacher, I used to regularly warn my students that “education dulls your brain: the more education you get, the duller you become.” They invariably looked back at me with compassion.¹

Still, it’s hard to ignore the statistical evidence pointing to the relationship between education attainment and higher lifelong income levels. But that’s not what this post is about. It’s called: “Is Education Good for You?” not “Will Education Make You Rich?”.

But there do seem to be non-financial benefits. Take criminal behavior. The Saskatchewan microsimulation suggested that fully reducing the educational attainment gap among Indigenous people could reduce annual population-wide contacts with police for criminal offences by as much as 15 percent. That’s a statistical projection rather than an observation of events, but it’s at least a useful data point.

On the other hand, the most public, brazen, and sustained criminal² activities of the past couple of years across Canada have occurred at on-campus demonstrations. Those students (and those of their professors who stood at their side) might be educated, but their education doesn’t seem to be leading them to the right end of the criminal justice system.

And - without implying a direct comparison - it’s been well documented that many of Hitler’s Einsatzgruppen commanders - the guys whose jobs involved the face-to-face shooting of hundreds of thousands of Polish and Russian Jews beginning in 1941 - were not just educated, but often held multiple doctorate degrees.

So education certainly doesn’t guarantee decent behavior. But does it make decent behavior more likely?

For argument's sake, let’s take the Saskatchewan microsimulation claim at face value and assume that there is a measurable positive correlation between more education and less crime. But is there any hard evidence that it's the education itself that's making the difference? Perhaps people with the personal qualities necessary to succeed at education - like self-discipline and innate academic ability (however that might be measured) - are already less likely to be attracted to crime.

Assuming we could show that there's something about the curriculum itself that steers students away from crime, it’s worth trying to understand whether the intensity of that effect fluctuates in sync with changes to K-12 and higher education curricula.

To be sure, there is reasonably convincing evidence for long-term causal benefits from Social and Emotional Learning (SEL) and early intervention programs. But many of those happen in K-8 environments, so they tell us little about the social value of secondary and post-secondary education. And even in the context of the K-8 world, such programs are generally applied as focused interventions to address existing problems for specific students.

I’d be interested to know whether there’s evidence that one or more provincial curriculum frameworks that public school secondary students are exposed to have a measurable positive causal impact. To the best of my knowledge, this hasn’t been formally studied.

And that’s a shame, because you’d think that decisions involving the billions of dollars we spend implementing public education curricula would have been informed by evidence. After all, the very lives of our nation’s children are at stake.

Designing and executing such a study shouldn’t be all that difficult. You might decide to focus on individuals who graduated from high schools in multiple provinces between, say, 2005 and 2015 and identify as many curriculum variations as practical. Alternatively, you could track single-province outcomes before and after major changes to curriculum objectives. Either approach would allow you to compare programs focusing on things like civic education, social justice, vocational training, financial literacy, conflict resolution, STEM, and humanities. Those could then be correlated with real-world criminal justice trends.

Of course, you’ll need to control for as many confounding variables as possible including variations between curricula, school funding, demographics, socioeconomic status, and access to social services.

Crime rates are just one useful measurable outcome we could study. Here are a few more:

• Voter turnout rates

• Volunteer hours

• Charitable giving rates

• Rates of long-term relationship stability and marriage longevity

• Incidents of discrimination or bias-motivated actions

• Cross-cultural friendship patterns and intergroup contact

• Ethical Decision-Making Indicators

• Academic and professional integrity violations

• Responses to moral dilemma scenarios in longitudinal studies

• Rates of substance abuse and addictive behaviors

I get it that most of the people who normally do this kind of research are probably more interested in testing new and ground-breaking educational theories with an eye to replacing the stuff we’re using right now. But to be blunt, most of those new and ground breaking theories are really just rehashes of what was fashionable 10, 20 and even 60 years ago (Hall-Dennis, anyone?). Wouldn’t it make more sense to use available data to figure out what’s actually worked in the past?

1. Of course, I also used to tell them that “if ignorance is bliss, this would be the happiest place on Earth”.

2. At least assuming Criminal Code sections 423 and 430 were still on the books.