Mark Perry posts regarding the new AEI Education Outlook by University of Missouri economist Cory Koedel which shows Education to be by far the easiest course of study in most colleges. Mark finds additional evidence from Cornell University to back up Koedel’s claim. Education majors enter college with lower SAT scores than students majoring in other fields but leave college with higher GPAs. Unless something truly magical happens in education schools, we can only conclude that Education is simply a less demanding course of study. As Koedel points out, this lack of rigor undermines rewards to students who work harder, makes it more difficult for schools to distinguish good teaching candidates from poor ones, and may contribute to a professional culture that cares little about standards of quality.
But, as a forthcoming paper that I have co-authored with Jason Richwine will show, the low standards applied in education degrees also complicate the task of determining whether public school teachers are fairly paid. Teachers claim to be underpaid because they receive lower average salaries than private sector workers with similar levels of education. (Our paper shows that, even if this is true, they more than make up the gap through generous benefits, but we’ll ignore that for now.) But note that the control variable here is the level of education — meaning, Bachelor’s degree, Master’s degree, and so on — and not the quality of education nor, more importantly, the ability or productivity of the worker.
In most public-private pay comparisons we can use the level of educational attainment as a proxy for individual ability (For instance, see our paper on federal pay). This isn’t because every college major has the same level of difficulty or that every college provides the same quality of education. It’s because a large number of different college majors are distributed across a large number of different occupations, so given an adequate sample size the inadequacies of education as a proxy for productivity wash out. The fit of the estimate isn’t as tight as if you had some better measure of employees’ ability (meaning, in stat-speak, that the R-squared value of the regression isn’t as high) but the results aren’t biased in one direction or the other.
But when examining teacher pay the problems with educational level don’t wash out: most teachers have degrees in education and most people with degrees in education work as teachers, so if an education degree signals lower ability or a less rigorous education — and Koedel’s paper and other sources indicate that’s the case — then regressions using education as a control variable may falsely show public school teachers to be underpaid.
Put bluntly, public school teachers enter college with below-average SAT scores, major in the easiest undergraduate course of study, take Master’s degrees in education that have no appreciable impact on teaching quality, and then wonder why they’re not as well paid as someone who got a Master’s in chemical engineering. They shouldn’t.