Benjamin Studebaker

There is much discussion of evaluating teachers these days. Recently, the Chicago teachers union went on strike over the issue, among several others. The premise behind teacher evaluations–that the quality of a teacher can be determined by standardised test results, is rather tenuous. Today I would like to discuss some of the issues with this method and propose a superior alternative.

When I was in high school, I had a particularly poor teacher for one of my subjects. One day, this teacher told us that one of the assistant principals was going to evaluate said teacher. I thought it might lead somewhere, but I had my doubts. This teacher had been at the school for an awfully long time. Surely there had been past evaluations before that somehow this teacher had managed to survive. When the day of the evaluation came, I was surprised to see that the class period went totally unlike any other day of class I had ever had from this teacher. There were none of the usual back and forth arguments between the teacher and the students about the poor quality of the assignments, problems with the scheduling of tests and quizzes, or people pointing out that hey, we were not learning much of anything. Instead, the teacher arranged for us to play a game all class long, a game that was reasonably entertaining. We did not do any work, there were no opportunities to expose the usual flaws in the teacher’s planning and teaching. Sure enough, a few days later I was delivering mail to some teachers from the office and ran across this teacher’s evaluation from the assistant principal. It was positively glowing, full of praise. This taught me something:

If the teacher knows that the teacher is being evaluated, the evaluation will fail.

So what about evaluations that are based on test scores? A teacher can fake through an evaluation based on an anecdotal experience, you cannot fake test scores, at least not without cheating. This is true, but there are nonetheless several problems with this:

• Problem of Insufficient Controls: there are many factors that influence student performance aside from teaching that cannot be controlled or accounted for in these evaluations
• Problem of Teaching for the Test: while it is theoretically possible to write a test that actually does contain everything that a student should learn in a given class in a given year, many standardised tests are overly specific in methodology and are consequently exclusionary
• Problem of Grade Inflation: in countries with heavy emphasis on using testing to analyse teacher or school quality, grades have a tendency to rise not because students know more but because tests become simpler

The first two issues were more directly referenced by the Chicago Teachers Union, the latter has been drawn to my attention during the course of my travels in Europe.

Insufficient Controls:

This is the most obvious problem–teachers have limited efficacy in raising student scores. Teaching is one of a great many factors involved–parenting quality has a huge role to play. Poor parents tend to have less time to be involved or to be less involved by disposition. Absentee parenting is more common among the poor, incidence of abuse, alcoholism, drugs and all sorts of other negative influences are stronger in poor areas, and as inequality has grown over the last thirty years, the degree of alienation in this neighbourhoods has only exacerbated further. Poor people tend not to be able to afford to live in middle or upper class neighbourhoods, and since school districts typically allocate students geographically, this means some schools are just absolutely full of poor students while others are nearly devoid of them. The propagation of private schools and vouchers only serve to allow poor parents who do care and are involved to send their children elsewhere, further diminishing the level of parental participation in the learning process at these schools. Even a great teacher would struggle to post consistently high test scores in such an environment. To continue our Chicago theme, it’s like expecting Michael Jordan to go play basketball for the worst team in the league and still expect him to win the league championship. A great teacher needs great parents like a great athlete needs great team-mates. Teaching with bad parenting just cannot produce the result that teaching with good parenting can, just as Jordan playing with Brian Scalabrine would never bring about the same result as Jordan playing with Scottie Pippen.

Teaching for the Test:

It is often argued that tests can be written to encompass everything a given student should learn in a given class in a given year, but empirically the tests do not tend to be quite that good. Speaking to one of my former teachers recently, I learned that in one year’s English class, students are now writing the same kind of essays stylistically all year long because the standardised test they take at the end of that year emphasises this style and the school is seeking to raise the pass plus percentage. After that, the teachers are so sick of grading and teaching that style of writing that they never focus on it again for the remainder of the students’ time in high school. Having myself taken tests like the SAT and GRE, I can testify from experience that there really is a certain manner in which these tests are written that heavily advantages students who understand the language of standardised tests intuitively or have a great deal of experience in interpreting it. That is not written out of bitterness–I usually find myself personally advantaged by the system of standardised tests. In the course of the effort to get all students to pass these tests, there is a loss of opportunity for creative thinking by both students and teachers, the latter with respect to lesson planning, the former with respect to the kind of work done in school, and students with less exam aptitude are passed over or looked down upon by universities and teachers, damaging their self-confidence and self-respect.

Grade Inflation:

In the United Kingdom, a curious thing has happened since the introduction of exam-based teacher and school assessment–scores have risen, but the education system’s performance compared with other countries has simultaneously declined. The cause of this is grade inflation–test-making companies have discovered that schools and teachers will choose the easiest tests so as to most easily achieve their targets and pass their evaluations. As a result, tests in Britain have grown steadily less challenging over time. This illustrates that not only are exam-based teacher and school evaluations ineffective at discerning teacher performance and result in less creative education, but that it directly leads to the dumbing down of the evaluations themselves until they no longer serve their intended purpose at all. If everyone passes the exam, the exam teaches us absolutely nothing about anyone’s abilities relative to anyone else. It does not discern levels of aptitude and does us very little good.

How should we evaluate teachers instead? Let the students evaluate them. Give every student the opportunity to assess the job performance of his or her teacher. Have every student sign his or her name to the evaluation, to encourage it to be taken seriously. This would empower students and make them feel directly involved in the education process, rather than simply victims of it. Some people might say that students would not take the evaluations seriously, or would deliberately slander teachers they just were not fond of. Doubtless, this would happen, but it would be in the minority of cases. Most students actually do have some interest in being given the necessary tools to do well, or the very least, mentally engaged. The principal can read these evaluations, and sort out or dismiss the ones that do not take the task seriously or are written by students with persistent reputations of delinquency. No longer would students be victims of bad teachers they had no hand in evaluating, and being treated like mature adults rather than criminals might just more respect and positivity in relationships between students and their schools.

I certainly know that, if the teacher I discussed above were evaluated in this manner, there is no way that teacher would still be around.