Early last month, NPR reported that 178 teachers and principals in the Atlanta school system fudged student test results. That is, they cheated. I’ve been meaning to comment on this for a while, but have been busy with classes, and finally have a little free time this week.
What I find most surprising about how this story has been reported is how shocked everyone seems by it. The current system of assessment (which is a direct result of the No Child Left Behind (NCLB) legislation) creates an incentive system which rewards teachers that cheat. The system is not designed to reward quality educators, but rather to reward educators whose students manage to turn in better test results. It is a broken incentive system, and it is no wonder that people are trying to game it. I am surprised that we haven’t heard more stories like this!
The sad fact is that teachers are currently treated as interchangeable cogs in a machine. In my experience, limited though it may be, the work of a modern primary or secondary instructor is incredibly structured. Large swathes of the teaching calendar are given over to state- and federally-mandated standardized tests, and the curriculum is handed down from above with little-to-no input from the people that actually have to teach it.
Even worse, when these standardized tests and curricula fail to achieve the desired results, schools are tied down with even more restrictive and useless instructional materials, teachers can lose out on pay or even lose their jobs, and students flee for other schools, leading to massive losses in funding.
In short, it is teachers (and, to a lesser extent, principals and other low-level administrators) that are blamed when schools fail to meet adequate yearly progress as defined by the legislators responsible for NCLB, yet these same teachers are not given the freedom to teach in the manner they might consider most effective, and are judged not on the basis of any direct measure, but on how their students perform on standardized tests. Good teachers are those with students that pass, and bad teachers are those with students that fail.
This is a broken, broken system which was designed, from the ground up, to fail.
By way of analogy, consider a researcher who is interested in the population of blue whales in the Pacific Ocean. Blue whales swim deep, and are difficult to observe directly, so this researcher proposes to monitor krill populations instead. Blue whales eat krill, and krill are easy to observe, so it seems reasonable to assume that one might understand whale populations by proxy.
There are obvious problems with this approach. Blue whales are not the only factor which control krill populations. Other whales eat krill, and changes in currents or weather could have effects on krill populations. It may be possible to account for many of these factors, and there may be some useful data about blue whale populations that can be learned, but it is a pretty silly experimental design overall because it measures the wrong thing.
If you want to know about blue whale populations, the best approach is to find a way to count blue whales. Measuring by proxy may be useful, but it is likely to be a limited method.
Similarly, if you want to determine if a teacher is a good teacher or a bad teacher, the best approach would probably be to directly observe that teacher. Instead, we look at student test results, and assume that these relate directly to the quality of the teacher. This is like trying to count blue whales by counting krill. In addition to the influence that any one teacher may have on student performance, there are innumerable other variables that can cause students to perform well or poorly. The wealth and first language of students matter, as do other socio-economic factors. The quality of past instructors is important. The level of parental involvement is a huge factor. Last year, while I was teaching middle school math, we received results from a set of standardized tests in which students in the last third of the alphabet performed significantly worse, as a group, than the average. It turned out that this group of students had been assigned to modular classroom with a very loud A/C unit. It seems obvious to me that the influence of one teacher on test scores can be easily swamped by other factors.
Thus the quality of instructors is measured by proxy, and the proxy is clearly a poor one. So, to return to the vital question, is it any wonder that teachers are cheating?
And now for the bit that I haven’t seen discussed much: teachers are professionals. They have professional training which is meant to qualify them to teach in the same way that medical school qualifies physicians to practice medicine or lawyers to practice law. Yet even armed with this training, nearly every parent, politician, and member of the peanut gallery seems to think that he or she knows what is best for education.
This is, perhaps, understandable. Everyone goes through some schooling, and everyone comes out the other end thinking that they know something about teaching and learning. This is a nearly universal experience. However, it is an experience that is had by young people who have little to measure it against, and who quickly form uninformed opinions. I mean, teachers work short hours and get the summer off—it can’t be that hard, right?
By contrast, physicians and lawyers have access to knowledge that seems arcane to most outsiders. They use language that is obscure, and have their own professional organizations (medical boards and legal bars) which govern their behaviour. There is a mystery and mystique associated with these professions that teaching lacks.
The reality is that teaching is not an easy job. I know of very few teachers who actually take off much time during the summer, and I know of even fewer teachers who actually go home when the bell rings. Most teachers that I know put in 10-12 hours per day, 7 days a week during the school year, and continue to work hard throughout the summer. Like physicians and lawyers, teachers are required to continually take classes to develop and expand their skills, and they are required to meet other professional obligations.
Yet unlike physicians and lawyers, who police themselves through medical boards, legal ethics councils, and the like, teachers are governed by non-teachers. In the recent debate about Obama’s health care plan, fear was raised about the possibility of “government death panels”: committees of government appointed bureaucrats with no medical knowledge who would make decisions regarding an individual’s treatment. Despite the absurdity of the claim, this was a frightening prospect for many, and caused vocal outcry. Why, then, are we not concerned with committees of bureaucrats with no expertise in teaching and learning making decisions regarding individual education? Where is the outcry here?
The problem is that we are asking the wrong people to sit in judgement over teachers, and they are choosing the wrong metrics. We are essentially asking uninformed outsiders to rule on whether or not specific teachers and schools are doing their job properly. Using student test scores to assess teacher quality is like counting the number of patients that die in the care of a particular physician to judge the quality of that physician. It might seem like a good proxy, but geriatric physicians are going to look terrible when compared to pediatricians.
If we want higher quality teachers and a better educational system, the first people that we should be talking to are teachers. Instead of treating them like assembly-line drones on the factory floor of education, why not treat them like the professionals that they are, and listen to their opinions regarding a field in which they have expertise?