Not a matter of degrees

by John Hood, Syndicated columnist and NC SPIN panelist, October 26, 2015.

For three years now, I’ve been defending the decision of the North Carolina General Assembly to end the longstanding practice of giving public-school teachers automatic pay boosts if they possess or obtain graduate degrees.

The irony has been striking. Critics have consistently argued that getting rid of the pay bumps reflected an anti-intellectual, anti-scientific mindset on the part of Republican lawmakers. But the intellectual case for paying teachers more for graduate degrees is nonexistent. One of the clearest, most replicated findings in the social science of education policy is that there is no meaningful relationship between graduate degrees and teacher effectiveness.

To put it more bluntly: those who continue to insist North Carolina should reinstate the pay bumps are either entirely ignorant about the subject or contemptuous of the very higher learning they claim to champion.

Since 1990, scholars have published more than 100 studies in academic journals that tested the relationship between teachers having graduate degrees and some measure of educational success, such as test-score gains or increases in graduation rates. In more than 80 percent of the studies, there was no statistically significant relationship. A few of the studies actually found a negative effect. Only 15 percent produced a positive association.

Just this year, two new working papers — not yet published in journals but already posted by reputable researchers working with North Carolina data — have confirmed the basic findings of prior studies while using innovative new designs. One of them was co-authored by Duke University professor Helen Ladd, a vociferous critic of the education policies adopted by Republicans in Raleigh.

Ladd had previously published several studies with co-authors Charles Clotfelter and Jacob Vigdor that found no evidence for graduate degrees making teachers more effective in the classroom. After the General Assembly cited this research as justification for eliminating the pay bumps in 2013, she decided to take another look. To her credit, in her new paper Ladd candidly admits that she expected a refined research design, focused on middle- and high-school teachers and controlling for potential selection bias, might reverse her previous findings and demonstrate that graduate degrees do, in fact, make teachers better.

The expected result didn’t occur. In their new paper, posted by the National Center for Analysis of Longitudinal Data in Education Research (CALDER), Ladd and co-author Lucy Sorensen found no effect on student achievement. They did find a modest effect on absentee rates in middle school, which is pretty weak tea given the financial and policy implications.

As an identical twin, I admit to a strong personal interest in another study released on the subject this year, by researchers Moiz Bhai and Irina Horoi from the University of Illinois at Chicago. They also examined North Carolina data, but with a twist: they identified sets of identical twins and then made use of the fact that some twins sat in the same classrooms while others were assigned to different teachers. Twin studies are valuable because they automatically adjust for natural and environmental factors that other studies may have trouble identifying.

In their interesting paper, Bhai and Horoi confirmed earlier research findings: that there are significant but diminishing returns to teacher experience, for example, and that National Board Certification may be a good way to identify high-performing teachers. As for graduate degrees, they once again found no consistent relationship to student achievement.

To conclude that teacher pay supplements for graduate degrees are a waste of tax dollars is not to conclude that higher learning is useless. There are plenty of professions for which advanced degrees are clearly valuable or even required. And teachers seeking to become school administrators, as my father did during his career in the Charlotte-Mecklenburg system, might well benefit from additional education — particularly if North Carolina colleges and universities reform their principal-training programs as the education-policy group BEST NC recommends.

As for those who still fume about the legislature’s 2013 decision to end the pay bumps and use the money to boost starting pay and fund large early-career raises, however, I find their disdain for scholarly research puzzling.

http://www.carolinajournal.com/daily_journal/index.html