Teachers paid 19% less than comparably educated peers

From the New York Times

by Jeffrey Keefe

The authors of "Assessing the Compensation of Public-School Teachers" conclude that teachers receive total compensation 52 percent greater than fair market levels, which translates into more than a $120 billion overcharge to taxpayers each year. There are numerous and significant problems with this study. I will examine one of them, wages.

The study begins by comparing teachers to other workers with similar education, experience and weekly work hours, and using this model the authors find that teachers are underpaid by about 19 percent. The study then questions its initial finding by raising doubts about the rigor of an undergraduate degree in education and the average cognitive ability of teachers.

Using the National Longitudinal Survey of Youth, the authors look at how much teachers earn in comparison with how well they score on the Armed Forces Qualifying Test, treating this as a measure of I.Q. In this model, teachers’ salary penalty becomes statistically insignificant, which would indicate that teachers are not underpaid. The study incorrectly concludes that eliminating education as a control variable and letting A.F.Q.T. alone account for skills provides the most accurate wage estimates, and the "wage gap between teachers and non-teachers disappears when both groups are matched on an objective measure of cognitive ability rather than on years of education."

There are numerous statistical problems with this model, and the conclusions drawn from it are erroneous. A cognitive ability model that does not account for education level is meaningless, because individuals are employed in jobs that depend on the skills acquired through education. This is why education level, not cognitive ability, is used to calculate human capital: it captures the level of investment in workers’ skills.

While cognitive ability may be necessary to acquire an education, it does not in any way indicate that an individual has made a personal investment. There are people with high cognitive ability who have not completed high school who are performing unskilled work, while there are people of average cognitive ability who are performing highly skilled jobs. Comparing teachers to people of the same A.F.Q.T. level who have not invested in their education does not demonstrate whether there is or is not a teacher wage penalty. In other words, cognitive ability may enable educational investments; however, it is education that creates the skills that facilitate professional performance. Consequently, measured cognitive ability is correlated with wages but explains little of the variance in wages across individuals and time. The only reliable comparison in this report is its starting point: there is a 19 percent wage penalty for teachers.

http://www.nytimes.com/roomfordebate/2012/01/02/are-teachers-overpaid/a-better-way-to-slice-the-data