Michelle Kerr just finished her first year teaching, as a second-career intern. She has some interesting ideas that she shared recently in the Washington Post, (hat tip to Teaching Now). In reflecting on how teachers are paid, she rejects the idea that we should be rewarded for additional education or longevity, but she finds some real problems with merit pay based on test scores. She offers some changes that she believes would make this practice more acceptable:
- Teachers be assessed based on only those students with 90 percent or higher attendance.
- Teachers be allowed to remove disruptive students from their classroom on a day-to-day basis.
- Students who don’t achieve “basic” proficiency in a state test be prohibited from moving forward to the next class in the progression.
- That teachers be assessed on student improvement, not an absolute standard -- the so-called value-added assessment.
First of all, kudos to Ms. Kerr for getting out there with her views on issues affecting her new profession. If every teacher were working on solutions and sharing their views in print we would have a much richer education dialogue.
She also focuses attention on four areas where current merit pay proposals are manifestly unfair. Teachers in schools with attendance and discipline problems face huge obstacles in raising student achievement levels. There are also high levels of transience at many schools. And as she points out it is unfair to hold a teacher accountable for teaching Algebra if you load the class with students who have not mastered fractions.
But I want to take a closer look at the suggestions she offers, not in regards to their effect on teachers, but focusing instead on how they affect students. And let’s recall that incentives can have perverse effects.
First of all, attendance. Imagine we implement the suggestion that only students with 90% or better attendance count towards my merit pay. Will low-performing students find themselves marked absent with greater frequency? Will tardies be treated more harshly? Since teachers are responsible for taking attendance, this seems like it could be a very tricky area, susceptible to corruption. It would be very tempting to bias the pool of students on which one is judged by shading the attendance in this way.
Second, she suggests that teachers be allowed to “remove disruptive students.” Teachers already have this option. What happens of we enshrine this as a right, and connect it to teacher pay? How much quicker will students with minor behavior issues find themselves languishing in some educational purgatory, with all the other randomly shaped pegs that refuse to fit in the square holes we are insisting on? This is likely to increase the rate at which students are “pushed out” of our schools, and put huge pressure on our site administrators and alternative schools to handle all the pushouts.
Third, along similar lines, we have the idea that we stop promoting students based on age, and instead make sure they have mastered all the pre-requisite content before moving on. This is a complex issue. Demanding that we must have well-prepared students before we can be expected to do our jobs will lead us to reject those students who are behind. That does not serve these students well. The research shows that neither retention nor social promotion are an adequate response to this. These students need a combination of a challenging curriculum, effective teachers, and differentiated instruction that targets the needs of individual students. Tying our pay to their test scores is likely to lead, once again, to the desire to reject or push out students that are going to make us look bad because they are not able to reach the standards.
Lastly, Ms. Kerr offers value-added as a a valid alternative to the system where this year’s students are compared to last year’s crop. Unfortunately value added compensates for only one of the many variables that make these scores unreliable, as David Cohen described here.
The real problem with tying our pay to test scores is that these scores, even when compared using a value-added model, are an inadequate marker of that which we are responsible for teaching. When our pay is tied to these scores, we will find ways to make the scores go up - but that will not mean student learning has really improved. That is why scores on state achievement tests show steady improvement, while the National Assessment of Educational Progress scores have remained flat over the past decade.
The more we focus on all these incentives, and the means by which they might be made more “fair,” the farther we get from the real interests of our students, which are served by teachers whose first concern is their well-being as individuals. We need to be able to accept our students as they are, whether or not they are at grade level, and whether or not they are able to show a year’s growth on a multiple choice test. We do need support to cope with disruptive students, but we should not be primarily motivated by a fear that this will lower our scores -- or our pay -- but by a concern for all of our students’ well-being, including those with behavior issues. We need to focus on a rich curriculum that challenges our students to think critically, to communicate and collaborate - rather than simply answer questions in a multiple choice format. Once we accept that these tests can be an adequate measure of our students’ work we have given up on our own value as professionals to define what learning is all about. This will not serve our students well, even if we are able to raise their scores.
By the way, over at Teachers’ Letters to Obama, we are preparing for our first Teachers’ Roundtable, where we will hear the views of a panel of teachers from across the country, addressing the failings of our current test mania, and offering alternatives. The event will be held on Monday, June 28th, from 5:30 to 7:30 pm Pacific time, 8:30 to 10:30 pm Eastern time. Registration for the online forum here.
What do you think of Michelle Kerr’s suggestions for making merit pay more fair? Will this help our students?