It sure is a confusing time to try to follow education policy. Bill Gates is writing clearly and sensibly about why teacher data reports should not be released, and even more cogently about the kinds of things we need to put into place if we are going to use teacher evaluations to help students learn.
The new teacher evaluation deal reached last month between New York State education officials and the state teachers' union allows school districts to base up to 40 percent of a teacher’s annual review on student performance on state standardized tests. The deal prompted Michael Mulgrew, the United Federation of Teachers president, to say of some teachers: “If they can’t improve, that means they need to leave the profession ... I feel bad when this happens. I tell them, ‘I understand you have this passion inside you, but that doesn’t mean you have the ability to be an effective teacher.’ Teaching is a tough profession.”
The U.F.T.’s vice president, Leo Casey, is arguing against the New York Principals' paper, a document that thoughtfully disagrees with using student test scores to evaluate teachers. More than 1,000 principals in New York State have signed the letter.
Although the U.F.T.'s objections are surprising, perhaps those of us who signed the paper should be simply grateful for the attention: no one else involved in the decision-making process around how teachers in our state will be evaluated seems to notice or care that over 30 percent of the principals in the state believe that a major component of the new teacher evaluation system is terribly flawed.
My concern about the agreement is that a large portion of a teacher's evaluation is to be taken out of the hands of principals. I am disturbed by this, not just because I think this will lead to inaccurate ratings and will pressure teachers in unproductive ways (it will), but also because I believe it speaks to a growing distrust of or disrespect for principals. I am surprised that the teachers’ union would trade a principal's rating for that of a student's test score, especially given the recent teacher data report debacle. Are most principals less fair or trustworthy than reductive data? I think not. I think most principals feel exactly as Mr. Mulgrew does when they work with an ineffective teacher, and they communicate those concerns with the same intelligence, honesty and kindness Mr. Mulgrew expressed above.
The desire to use multiple measures to rate teachers seems like a smart idea. However, New York City's two experiments with value-added ratings in education, the teacher data reports and the school progress reports, have not produced reliable information. So far we have not discovered any measures which clearly correlate teacher performance to student learning. This new agreement will generate a teacher’s rating by using data which we know does not answer the question we are asking. Why? Are principals incapable of understanding data, incapable of interpreting it based upon what they see in their schools? I think not.
I think we can review our schools’ data in a much more nuanced and accurate way than any measure designed to encapsulate and compare the work of thousands of teachers working with hundreds of thousands of students. No less a prominent voice in this discussion, Mayor Michael R. Bloomberg, was recently quoted as saying: “The principals’ job is to decide who’s good, who’s bad. It’s their judgment; that’s their job.” Who could disagree?
But we principals, too, are part of the problem. Not because we have promoted the use of bad data to rate teachers, but because we may have allowed our attention to stray from our chief job of promoting professional growth, training staff, documenting teacher performance, creating community and defining what quality teaching and learning look like in our schools. Newly necessary distractions like marketing and fund-raising and data analysis may have seemed more important than getting into classrooms and working with teachers on how to plan lessons and ask questions. But if we let our attention waiver from those things which we know should be our primary focus, if we asked “How can we help students earn more credits?” instead of “How can we help students learn more?” then some of the distrust we see driving this new agreement is our fault, even if we believe that is what the school system and the general public wanted us to do. We may have felt less incentive to concentrate on the quality of classroom instruction in our schools because we are rated on other things, but we know our jobs. If we chose to focus on tasks outside of instruction, it makes sense that the void such a choice created was filled by psychometricians.
The movement toward a more rigorous teacher evaluation system is a good thing. But to do it well will require us to define clearly and unambiguously the things we value and the things we want teachers to include in their classrooms. Then we will need to determine how to measure those things, and those measures will need to account for the differences we see in our students. We are not there yet.
One thing is certain: the measures in the new agreement will be powerful drivers for the kind of instruction our students will receive, whether they account for 20 percent of a rating or 40 percent. Do we want the Global Regents exam, as it is now constituted, to be paramount in teachers' minds as they prepare lessons for our ninth graders? Or, do we value critical thinking and analysis and problem solving and the ability to ask questions? If we value the latter, shouldn't we evaluate teachers on their effectiveness in promoting those qualities in their students? Since no one would argue that the Global Regents assesses students’ abilities in those areas, why would we use data from such an exam to evaluate our teachers, even if only for a portion of their rating?
Tying teacher evaluations to such data seems more about political expediency than about educating students well, and leads us down a dangerous road. Paul Krugman, parsing the failure to appropriately address the fiscal crisis in Europe, writes that "matters were made far worse than necessary by the way Europe’s leaders, and more broadly its policy elite, substituted moralizing for analysis, fantasies for the lessons of history." I wonder if we are not doing exactly the same thing. If we are not addressing the issue of teacher quality head on but are substituting the political agendas of both the left and the right in place of what we can verify as true, we will hurt our students terribly. And no one will provide our students with a bailout.