While there is no shortage of opinions and controversy about how best to improve schooling, virtually everyone agrees that the quality of teaching is the critical variable.

To improve schooling is only to contribute to the future success of students, not to guarantee it. Formal education is only one influence on child development. Kids who grow up in single-parent families, in poverty and in unsafe neighborhoods — conditions that are highly correlated with each other, by the way — have disadvantages that won’t necessarily be dispelled by access to good schools.

That’s no argument for inaction, however, or for changing the focus to policy issues other than education. For example, government policy may have played a role in disrupting the traditional two-parent family, as conservatives persuasively argue, but it’s far from clear what government policies could reverse the damage.

The same constraint doesn’t apply to education. There is an emerging body of compelling empirical evidence suggesting that teacher quality significantly affects student performance and that certain qualities or policies are more likely than others to improve the quality of teaching.

In a new paper published in the Journal of Economic Surveys, a team of Dutch scholars analyzed the academic research on teacher quality conducted since the 1970s by researchers across the developed world. The authors picked only high-quality studies, excluding those with inadequate statistical controls or other defects. Then they summarized the results.

One of them will be familiar to readers of this column: teachers with graduate degrees are no more effective than teachers without them. This is one of the most replicated findings in modern education research — which makes it all the more outrageous when the North Carolina legislature is attacked for getting rid of teacher bonuses for acquiring graduate degrees.

Naturally, there were clear beneficiaries from the practice, which was in place for decades before the General Assembly ended it in 2013. Most of them worked in North Carolina schools of education. Paying teachers to get degrees produced a steady stream of revenue to these schools, even as it produced no detectable improvement in student learning, because most of the advanced degrees teachers acquired were in education, not in particular subjects. 

As it happens, the Dutch researchers did find at least some evidence that math and science teachers with graduate degrees in those fields — and only those fields — are more effective than peers who completed their educations as undergraduates. Similarly, while most studies of teacher certification find no differences in effectiveness between teachers who obtain certification through the traditional university route and those who obtain alternative certification, having a credential for math instruction is a notable exception.

The findings on certification, by the way, might come as a surprise to those who’ve read about North Carolina students faring worse when taught by teachers with alternative certifications. Keep in mind that those reports did not, by and large, attempt to adjust for other differences between the teacher populations. In other words, those who pursue alternative certification differ from other teachers in ways that have nothing to do with the certification process. If you don’t adjust for demographics and other factors, you can’t draw meaningful conclusions.

When it comes to teacher experience, most studies find a link between experience and effectiveness, but it’s not necessarily a linear relationship. That is, the relationship can’t be depicted with a straight line on a graph. The Dutch team concluded that “most studies find significant learning gains for the first couple of years of experience, but hardly any later on in the teacher’s career,” although there are a few counterexamples in the literature.

What does appear to matter, then? Based on the studies to date, the subject-matter knowledge of teachers, their performance while in college, and the rigor of the colleges they attend are important considerations. North Carolina policymakers should structure our recruitment, compensation and retention policies accordingly.

All of which assumes, perhaps unrealistically, that we can debate the issue of teacher quality on the basis of evidence, not politics or specialinterest pleading.