Musings on Performance Pay

As Washington State is considering changing laws to qualify for "Race to the Top" federal funds, the implementation of performance pay has become part of the education reform equation here and elsewhere.

Teacher accountability is a huge issue that will not go away. Performance pay seems to be a way to address this, but it is misguided, in part, for the following reasons:

1) If merit pay is based on absolute student performance, an experienced teacher in a struggling school would make less than a first year teacher who is blessed to get a job in an already high performing, high achieving school or district.

2) Within a transitory population, state and national test scores make no distinction between students who have been in the same school/district for several years and those who arrived only months or weeks before.

3) Within an at-risk population, test scores make no distinction between students who attend regularly, and those who have high levels of absenteeism due to inadequate supervision at home or suspensions/expulsions due to fighting, drug use, or any number of other challenging behaviors.

4) Within a diverse population, test scores make no distinction between students who have been ESL/ELL for one month, one year, or many years. With the science test, there is no "developmentally appropriate" option for students with special needs. All students are required to take the same test, and their scores may be due more to their difficulty in understanding the questions, and less to their content knowledge.

5) Performance measured by student gains within a school year seems far more equitable in the context of merit pay, but these results will also be skewed by absenteeism and lack of student motivation. Even some students who attend daily make a conscious willful effort to disrupt the learning environment for other students, and not all teachers are given the support they need to effectively address these problems.

6) Performance pay may create a culture of secrecy and isolation rather than collaboration, as teachers attempt to have their test scores higher than those of their colleagues. There will also be the phenomenon of "dumping" certain categories of students on some teachers rather than others, which already happens often with newer teachers because experienced teachers feel that they have earned the "right" to not have to deal with the low level classes or the students with serious behavior problems. Hence our least experienced teachers are being loaded up with our greatest teaching challenges. This also happens with more experienced teachers who have shown an aptitude for being successful with Special Education or ELL students, so the more effective teachers may ultimately end up with large numbers of students who do not perform as well on standardized tests.

That said, however, it is time that we say yes to teacher accountability as long as we can find ways to measure teacher performance in a manner that is transparent, fair, unambiguous, and reasonable. This shouldn't be a way to get a higher salary or to get paid more than our colleagues. If teachers are not doing their jobs adequately, we need to help them find ways to be more effective teachers, or we need to help them leave the profession. It seems absurd to say that we will pay good teachers more but allow bad teachers to stay in the classroom with a lower rate of pay. I don't want my kids in a classroom with an ineffective teacher, and I can't imagine anyone else does either, regardless of whether that teacher is being "punished" by not getting "merit" pay.

Post Script:  I have been reading about VAA's (Value Added Assessments) that are being used in districts across the country. In some cases they are being used as the sole indicator of a teacher's effectiveness, and some news outlets are actually using this assessment data alone to rank teachers from best to worst. No test can possibly measure all that a student has learned in one year. Also, a teacher's success with some students often depends on the amount of support that the teacher receives from other departments (especially Special Education and ELL), administrators and parents. In classes that attract highly knowledgeable, highly capable students, the test scores at the end may not be much higher than the test scores at the beginning if the test focuses on regurgitation of factoids rather than a development of thinking, process and communication skills.

I would be in favor of incorporating value added assessments as one component of teacher evaluation if the assessments themselves were valid. In the realm of science teaching, research has consistently shown that we should be teaching less content, going more in-depth, and developing critical thinking skills. Yet all of the assessments that are used are "a mile wide and an inch deep" and are rarely aligned with our curricular goals. I have yet to see any one assessment (even one I've developed myself based on my classroom curriculum) that can provide an accurate portrayal of what my students have learned in a semester.

I acknowledge that the use of student test data to evaluate teachers may be an unstoppable train. If that's the case, I believe value added assessments that can show learning gains are more fair than absolute test data (and I do not at all advocate anything like the bizarre value added formulas used in New York). But a teacher evaluation should include multiple classroom observations by administrators and master teachers, as well as an examination of attendance data, behavioral concerns, and student work over the course of a semester or year.