In the past few years, there has been so much focus on arguments about the use of student outcome data to help evaluate teachers and the role of evaluation in personnel decisions that there has been too little attention paid to the ways teachers in different circumstances view the issues. So here comes Stephen Sawchuk with a story with a subplot on that in a discussion of the District of Columbia's evaluation system for teachers.
The core of the issue is that there is no unique way to parse out the effects of different circumstances on student outcomes or even the classroom environment of teachers, and any way that you treat the issue is going to leave someone worse off and having a plausible claim to being treated unfairly. The existence of growth/value-added modeling doesn't eliminate the sensitivity of models to various assumptions, even apart from the fragility of growth modeling in general. Do you assume equal growth for all students in a year as a normative expectation (one of the assumptions of Bill Sanders), or should low-performing students be expected to gain more or less in a year? Step away from student outcomes and you have other problems: should teachers act pretty much all alike, should some teachers be expected to exert charismatic authority based on the behavior of their students or expectations of parents, and should teachers be expected to "fit" the culture of a particular school? You can't say yes to all three questions. And that's without considering different subjects or grade-level expectations, or the proportion of students with disabilities or who are English language learners.
The consequence of this dilemma is that you will see teacher concerns about fairness even when there is a solid working relationship between a union and management and an agreement to experiment with new evaluation systems and performance pay. I have heard of such rough spots from several teachers in Florida in districts that have experimented with performance pay, and a disproportionate number of them are special educators whose students either have behavioral or significant cognitive challenges or both. One example of that was public, in terms of the uneven distribution of merit pay in Hillsborough County in the first year or two of that experimentation. The district and union publicly said that their challenge in implementation was to work through such difficulties, and it has not been an impediment to additional agreements and the winning of the Gates Foundation grant.
But the internal concerns among teachers about unequal opportunities and unequal conditions tied to evaluation are going to be a fact of life, they should not be papered over, and there is no perfect or obvious solution.
The matter of teacher performance sounds hopeless — all the better reason to switch to “3rd grade in a box” type education delivery systems — imagine how much taxpayer money that could be saved by sending every student’s parents the “boxed” version of their children’s education each year (DVDs, workbooks, videos, website logins, etc.) — no more paying for school buildings, teachers, utilities, insurance, schoolbuses, lunchrooms, etc. We are talking about many hundreds of billions of dollars through the coming century. Given that teacher performance cannot be measured and that the entire secondary school system is pedagogically and financially bankrupt, it is high time for society to discuss how to migrate K-12 out of the industrial-age version of schools into the information age sometime during the coming decades and certainly before the end of the 21st century. There’s no time to lose…