We need a Chase Mielke in every urban classroom. Let's get to making it happen. Photo courtesy of the Kalamazoo Gazette

One problem within the debates over education reform is what I call the “magic bullet-shooting holes” fallacy: The tendency of each side to either attempt to use some sort of magic bullet to prove their argument or tear down the argument by complaining that the research is full of holes. Given the fact that education research is, for the most part, so notoriously lacking in rigor that debates can end up being little more than shouting matches with five-dollar words in substitute for salty language, this isn’t surprising. But it often means that one of the two sides tend to miss the point entirely.

Stanford economist Eric Hanushek– one of the foremost researchers in education — exemplifies this in a commentary on Education Next about addressing the low level of academic instruction in America’s poorest schools. Arguing that there is more inference than evidence that low teacher quality is the underlying cause of woeful student achievement, Hanushek then declares that several of the key methods used by school reformers to determine this — most-notably the teacher salary comparisons pioneered by Marguerite Roza and the Education Trust — offer little evidence that this is so.

Certainly the Roza model isn’t exactly foolproof. Some of the worst-performing school districts certainly have plenty of veteran teachers. Which is often as much a problem in those districts as having far too many inexperienced teachers. Considering that just 1.4 percent of tenured teachers are ever dismissed for performance issues (and less than seven-tenths of one percent of newly-hired instructors are ever fired), the veteran status of teachers merely means they have avoided felonious activity and more-rigorous performance management. Additionally, as  Dan Goldhaber and Michael Hansen of the Center for Reinventing Public Education pointed out in their report, the average 25-year veteran is no more effective at improving student achievement than a teacher who has taught for four years.

But the Roza method does offer a  good starting point for measuring teacher quality among and within school districts. Why? Because the nature of the current teacher compensation system — in which teachers must earn years of seniority and numerous degrees before gaining high levels of salary and benefits — means that salary can be used to measure the number of newly-minted teachers in a school or district. Salary and experience are positively correlated (even if experience and teacher quality may not). As Hanushek concedes, there is correlation between the number of rookies on a teaching staff and the quality of instruction. I have used Roza’s basic method in my own work, most-notably in a 2006 editorial on improving teacher quality in Indiana’s poor urban schools.

Yet Hanushek fails to consider the fact that there are other ways of measuring teacher quality in urban schools which can stand scrutiny. This is something he should know quite well.

There are teacher absenteeism levels: For one, the higher the level of absenteeism, the more likely students are being taught by substitute teachers — who, no matter one’s views on credentials, are usually teaching out of field and thus providing lower-quality instruction; the measure may also show whether a large percentage of a teaching staff is coasting towards burnout. There is also the percentage of teachers with less than three-to-five years of experience; Hanushek already concedes that there is a correlation between number of rookie teachers and quality of classroom instruction.

Another is the percentage of teachers reassigned to new schools more than once every three years; this allows researchers to determine the percentage of teachers who are part of the notorious dance of the lemons that occurs between schools year after year. One could even use teacher test scores on such tests as the Praxis I — which is required in most states for initial certification — along with the percentage of teachers who have failed those tests and retake them for a second or third time.  As Katie Haycock of EdTrust (Hanushek’s foil in this debate) also points out, even the value-added assessment techniques Hanushek pioneered is offering new evidence that low-quality teaching is at the heart of urban school failure.

It is sad that Hanushek (and, to a lesser extent, Fordham Institute research czar Mike Petrilli) engage in the same sort of “magic bullet-shooting holes” argument that plagues so much of the education reform dialogue. Improving the quality of education for the poorest students requires high-quality reasoning and dialogue, along with high-quality research.