U.S. President Franklin D. Roosevelt once said, “Democracy cannot succeed unless those who express their choice are prepared to choose wisely. The real safeguard of democracy, therefore, is education.” The sentiment has since been echoed by many who argue that an educated populace is critical for maintaining democracy.
Indeed, studies have shown more-educated individuals are more likely to vote.
But how strong, really, is the link between education and participation in voting? And what could be causing it?
“One reason for [providing] public schooling is to create an educated citizenry that can participate in the political process,” says Jörg Spenkuch, an associate professor of managerial economics and decision sciences at Kellogg. “But to what extent thatholds—and then whether being educated actually makes people more likely to participate—that’s still an open question.”
To help answer those questions, Spenkuch and his colleagues, Ethan Kaplan of the University of Maryland and Cody Tuttle of the University of Texas at Austin, compared the voting habits of people whose birthday fell just before versus just after the typical fall cutoff date for elementary-school enrollment in the U.S.
The researchers’ analysis of census and voter-registration data showed that there were indeed small but measurable differences in the total amount of education these two groups of people completed—and in their likelihood to vote. On average, people who were born just before the cutoff date received more education and were more likely to vote.
Furthermore, they found that, for every additional year of education a person receives, their likelihood to vote increases by 3 percentage points.
“It was previously not known that people on either side of this [cutoff] line go to the polls at slightly different rates or have somewhat different party affiliations,” Spenkuch says. “These effects are not large, but they’re clearly there.”
A natural experiment
To investigate how education changes a person’s political engagement, Spenkuch and his colleagues needed a way to track the future voting habits of an enormous number of people who were nearly identical except for the total amount of education they received over their childhood and young adulthood.
So the researchers used a quirk of the U.S. education system to find their nearly matched populations. In many schools, a student must be five years old by September 1 to enroll in kindergarten. As such, a child born on August 31 will start their primary-school education a year earlier than one born on September 2.
Using U.S. census data, Spenkuch and his colleagues tested if this cutoff effect led to longer-term differences in the amount of education people receive—and whether this connection might be associated with people’s future political habits.
By matching census information on education levels with administrative data on voter turnout and political-party registration, Spenkuch and his coauthors were able to set up a natural experiment comparing the political behavior of millions of people whose birthdays fell near the enrollment cutoff.
“Our approach is very close to an idealized version of this research design, which is to compare people right on opposite sides of the line,” he says.
Extra education matters
First, the researchers wanted to confirm that people born before the cutoff date do, in fact, go on to receive higher levels of education. Indeed, Spenkuch and his team found that these “early starters” went on to complete 0.034 more years of schooling on average than their counterparts over the course of their educational careers. They also found that early starters were about 0.5 percentage points more likely to graduate from high school.
That doesn’t sound like much, but “another way to interpret it is that there’s a small number of people—about three out of every hundred—who complete an entire extra year of schooling on account of having been born just a few days before, rather than after, the school entry cutoff,” Spenkuch explains.
And that tiny-sounding increase in total education is significant across a sample of millions of people. “It’s part of the reason why having these massive datasets is so important,” he adds, “because only then can you actually isolate the signal relative to the noise.”
Next, the researchers compared the two groups’ future levels of political engagement. Among people who became registered voters, being an early school starter increased their turnout by a quarter (0.25) of a percentage point. It also decreased their likelihood of registering as either a Democrat or a Republican by about 0.18 percentage points.
Essentially, starting school as one of the youngest in a class—and attaining, on average, slightly more education over the next dozen years—resulted in more political engagement and less overt partisanship.
Quantity over quality
Then, Spenkuch and his collaborators analyzed the difference between the quantity and the quality of the education people received to better understand the forces driving the relationship between education and voting.
Consider two 50-year-olds who both completed college, but only one of them was an early starter. The quality of the two 50-year-olds’ education levels could be seen as roughly the same, because they both achieved a college degree. But when those two people were still in college, the exact difference in quantity between their educational attainment is still in play. At age 20, the early starter would be about a year ahead in their schooling.
By looking at the differences in turnout rates between people 19 to 21 years old on either side of the school-entry cutoff, the researchers were able to isolate how important the sheer quantity of schooling was to a person’s political activity.
“We show that every additional year of education increases voting turnout by about three percentage points,” Spenkuch says. “It’s a large effect.” (For comparison, research has shown that the introduction of television decreased voter turnout by less than two-tenths of a percentage point per year.)
The findings do not necessarily close the book on the researchers’ initial questions in one shot. But to Spenkuch, the implications of this research are clear.
“What it means is that providing people with more education makes them more likely to participate in the political process,” he says. “If you want to wind this back all the way to arguments for the public provision of education, it does seem to be the case that a more-educated citizenry is at least more engaged in politics. Hopefully, they’re also more qualified.”