Last week, Secretary of Education Arne Duncan gave an important speech at the annual convention of the National Education Association (“NEA”), one of the country’s two major teacher unions. Urging bold action, the Secretary challenged the NEA delegates to think differently about the role of unions. He characterized reform as a moving train, and invited the NEA to “get on the train,” by abandoning the outmoded model in which almost all employment rights and benefits are based on seniority; reforming teacher tenure so that the decision to award lifetime job protection is meaningful and so that poor performing teachers can be removed; and designing meaningful evaluation systems that include evidence of student learning. I think it’s fair to say that the NEA hasn’t quite jumped on the train yet. I hope they, as well as their often more forward thinking brethren at the AFT, are considering it.
It is, of course, inconceivable in most organizational settings outside education that proposals such as Secretary Duncan’s would be considered “bold action.” With the exception of educators and perhaps parts of city governments, almost everyone in America works in systems in which career progression rests on factors other than just seniority; most workers are used to high-stakes evaluations related to performance; and few are guaranteed a job for life after just showing up for two or three years. Yet the Secretary’s proposals are, in fact, “bold” in education, and they face significant opposition. What accounts for this? I believe that our ability to make progress depends in part on understanding educators’ real fears of the Secretary’s reform train.
First, we must acknowledge that asking unions to “board the train” will essentially come down to asking union members to cast a vote to make their jobs less secure. Human nature being what it is, I suspect this vote will come only when the rewards for doing so are persuasive, or the cost of not doing so is unacceptable. Some forms of train travel may have to be mandated through legislation. At minimum, we must acknowledge educators’ fears in order to gain union cooperation and to design and implement systems that can work. In my discussions with union leaders and members, they often express deep distrust that any system that purports to base employment decisions on merit—whether decisions regarding career progression, layoffs, performance, or compensation—can be fair. I believe this is a legitimate concern, but one that can be responsibly addressed.
There are basically two ways to discern merit in teaching: 1) observing classrooms and evaluating the teacher pursuant to a rubric that describes what good teaching looks like; and 2) calculating student learning growth over time, based on test scores, and comparing it to the growth that can be statistically predicted for the same students. Observation is the traditional evaluation method, in use in almost every district in some form or another. However, as a great new study by The New Teacher Project shows, the vast majority of teachers observed are rated above average and almost none are rated unsatisfactory. To put it as many teachers do, “evaluation is a joke”.
This need not be the case, of course. There are several widely used systems for rating teachers based on observing them teach, and some districts, especially smaller ones, have implemented such systems quite rigorously. This method, however, is very expensive to do well, since it takes significant training to ensure that all evaluators rate teachers similarly, often requires additional staff as evaluators, and requires a system of checks and balances to prevent evaluator bias or error.
I believe we must keep our common sense about us if we attach real consequences to observational systems. Many of us have had teachers who changed our lives. Yet when we reflect on the teachers who made a difference, we realize that the best ones ran the gamut of behaviors – and many of those who made the most difference did not fit the behavioral ideals of these observation-based systems. Personally, I am firmly agnostic about teaching style, so long as the style fits in with school culture and students are learning (although I grant that all good teachers share a few characteristics, such as setting high standards and engaging all students). In any event, I understand it when educators are skeptical of teacher evaluation systems based on classroom observations. To me, used alone, these systems are a little like rewarding those in sales jobs for making the sales pitch, without taking into account whether they actually sold anything.
Like the Secretary, I believe that student learning growth is a critical success metric, since student learning is our business. There are significant fairness concerns on that front, too, however. Most of us believe that it is only fair to ask educators to cause students to grow in knowledge. To compare one educator to another, it is important to take into account factors such as student poverty, prior knowledge, and other variables that affect learning. Although statistical systems that can do this have now been used and tested in many contexts, and I believe they can reliably be used to make broad performance distinctions in some grades, they are plainly scary. Most people don’t understand sophisticated statistical models, and the higher the stakes, the more nervous people get about a system they can’t understand. This, too, is human nature.
A good strategy, which has been successfully used in the Teacher Advancement Program and other systems that offer additional compensation to high performing teachers, may be to combine rigorous observation with measures of student learning growth. It also may be best to implement learning growth measures in stages: using them first to give teachers and administrators insight into opportunities for improving instruction; second as a factor in evaluations; and, finally (maybe) as a component of base compensation systems. So-called “merit pay” (as opposed to layering bonuses on top of the lock-step salary schedule) is the most divisive subject in education, and there are reasonable differences of opinion about how to make it effective in better serving students. This is an area where I hope we will see lots of experimentation by charter schools.