Over the past few weeks, I’ve read many dozens of news articles, editorials, and research studies about current “education reform” efforts.

Today, I realized that nearly all the “disputed” ideas involve accusations of perverse incentives.

Here are a number of specific topics or issues; for each, I’ll start with “shared beliefs,” and then address the proposed improvements for that topic or issue.

(1) Teacher Evaluation, Teacher Quality: We all believe that in our schools, there are some excellent teachers, some good teachers, some mediocre teachers, and some bad teachers.  We all agree that all teachers (including the excellent ones) could learn and improve their teaching skills, and we all agree that there are some current teachers who should stop teaching.

The “real controversy” isn’t whether teachers should be evaluated; it’s about how teachers should be evaluated, and what consequences should follow the evaluations.

(2) Standardized Tests: For every complex problem, of course, there is at least one obvious, simple, cheap solution that is absolutely wrong.  Here, the obvious, simple, cheap solution is to evaluate teachers based primarily on their students’ performance (or improvement) on standardized tests.

This strategy (the core of the NCLB and RTTT laws) creates “perverse incentives” for schools and teachers to change their curriculum in ways that we all agree are wrong:

• In thousands of elementary schools, teachers and students now spend nearly all their time on “reading and math” (more specifically, those aspects of math and reading which are “testable and frequently tested” on standardized tests, to the exclusion of more complex aspects of math and reading).  Forget about history, science, art, or music — for elementary schools and teachers, consequences only attach to math and reading test results. (Secondary teachers of history and science will need to rework their curriculum to deal with large numbers of students with limited experience in those subjects.)
• Perhaps more insidious, many schools pressure teachers to focus on “buffer students,” who are “just below” a proficiency level, to the exclusion of other students who need help, but aren’t likely to jump to the next proficiency level on the next test.
• In some schools (but not others), many students are “held back” to prevent them from being included in test results for the next grade level.
• Some schools allocate resources to find ways to exclude, transfer, or expel students with low test scores (but not students with high test scores).
• Some states “dumbed down” their content standards and their standardized tests, to make it easier for students to achieve “proficient” scores.
• Administrators may deliberately place specific students in schools or classrooms to improve or worsen that school’s or teacher’s test scores (thus, the system can easily be “rigged” against certain teachers, in favor of preferred [compliant] staff).
• Student performance is generally higher in elective or leveled classes where students “self-select,” creating a perverse incentive for secondary teachers to avoid working with low-performing students, when possible. Some teachers might also seek out elective classes for which there are currently no standardized tests.
• The “high stakes” have led some teachers and administrators to cheat, including correcting answers on student answer sheets before they’re submitted for grading.

It’s also important to recognize that these standardized tests were originally designed to evaluate student performance, not teacher quality.  And student performance is affected by many factors other than teacher quality or effectiveness. Perhaps more troubling, students face no consequences for their performance on these standardized tests, which are “high-stakes” only for teachers and schools.

(3) Relationships Among Teachers: We all believe that every teacher can benefit from the exchange of ideas with other teachers. (That doesn’t mean that we agree on the value or importance of specific types of exchanges, such as peer observations, collaboration meetings, or curriculum uniformity.)

But performance-based “merit pay” creates competition between teachers within a district or school.  This creates a perverse incentive against collaboration and cooperation among teachers.  Teachers who share ideas will be at a disadvantage, as other teachers use those shared ideas alongside their own “proprietary” ideas. In some cases, some unethical teachers might deliberately sabotage other teachers’ efforts — and be rewarded.

Competitive “merit pay” does create another incentive, which some might like: good teachers in low-performing schools would be more likely to receive merit pay increases than good teachers in better-performing schools (if you believe that the overall quality of teachers is higher at the “better” school, making the competition harder there).

(4) Evaluating Teachers Fairly is Complex (and Expensive): We all agree that teacher evaluation is important but very complex.  Many of us also believe that “we know good teaching when we see it.”

Education professionals agree that teacher evaluation should include significant classroom observation time, and that teachers would also benefit from observing other teachers.  Unfortunately, subjective evaluation can be unfairly “colored” by many factors, including teaching styles or the relationship between the teacher and evaluator, so a fair system would include multiple evaluators.  And since every teacher has some “off days,” evaluation should cover multiple class sessions over several weeks or months.

But time is money, and there is constant demand for administrators’ and teachers’ time for a variety of “urgent tasks.” It’s just not plausible to expect schools to fairly allocate staff time for meaningful evaluations. (Ask some teachers: during each year you’ve taught, how many hours were you actually observed by someone evaluating your work?)

Likewise, we all agree that teacher quality can be improved by effective collaboration and professional development — but fairly allocating time for these is implausible in many schools.

(5) Fairness of Evaluations: Teachers and school administrators are human.  Each of us has biases and preferences about who we’d like to work with, and who we’d prefer not to work with. We have good days and bad days; we have distractions. These skew our perceptions, and they may even skew our intentions.

Ask a few teachers, and you’ll soon hear lots of stories about unfair subjective evaluations of teachers.  Now imagine attaching additional consequences to evaluations, and not just for new teachers but for all teachers in a school. Think about the impact of evaluations if “tenure” were eliminated, and teachers could be terminated “at will.” Now add the burden on staff to evaluate every teacher, not just new teachers. Now add the loss of collective-bargaining rights to negotiate a fair evaluation process or dispute process.

(6) Financial Perverse Incentives:  Huge budget cuts are also shifting more weight to some decisions, separate from other “reform efforts.”  For example, in most schools, class size limits vary for some subjects, with certain “elective” classes (art, music, drama, PE) allowing more students per teacher, and some other classes must be smaller.

Many states and schools are raising class sizes and teacher-pupil ratios, in order to reduce costs, but this makes teachers’ jobs even more difficult.

(7) Public Education, School Choice, Private Schools, and Vouchers: Many of the most vocal “school reform advocates” believe that much of public education should be “privatized.” They want to replace public schools (and teacher unions) with private schools (with no unions), for a variety of reasons: to save money, to support religious schools, to shift political power, and more.

Nearly all charter schools (and all private schools) benefit from a “self-selection” advantage (passive students and parents never apply), accompanied by more liberal expulsion policies.  (We should be alarmed that even with this advantage, plus longer hours, students in charter schools don’t outperform their public counterparts.)

The “No Child Left Behind” law, enacted a decade ago, promised huge funding to help schools (which never materialized, loading new costs onto already-strained school budgets), but in fact its primary goal was to “prove” that public education was failing.

The NCLB law asserted a bizarre, impossible goal: that schools should improve their test scores each year until 2014, when 100% of students were expected to score “proficient.” That’s a great aspiration, but not a reasonable goal (the simplest examples: non-English-speaking students who move to the US, special-needs students, and non-cooperative/resistant students).  Each year, more draconian penalties apply to schools that don’t meet this impossible goal — creating uncertainty and chaos which amplified the “perverse incentives.”

Now What?

If you read this far expecting a solution, I’m sorry to disappoint you.

We all agree that there’s room for improvement in education. We all agree that teachers would benefit from improved evaluation and feedback. We certainly don’t agree on the weight (if any) that standardized testing should be given in teacher evaluations.

Most of us probably agree that more resources should be allocated for classroom observations, meaningful collaboration, and high-quality professional development.

I doubt that many education professionals expect any of these things to happen soon.

Comments are closed