It’s time to admit that the idea of evaluating and paying individual teachers based on their students’ test scores is a loser. This logical-sounding strategy for improving teaching and learning sinks for multiple reasons: practical (standardized-test results arrive months after teachers are evaluated each spring); psychometric (these tests aren’t valid for one-shot assessments of individual teachers, and it takes at least three years of value-added data for reliable patterns to emerge); staff dynamics (when individuals are rewarded, collaboration suffers); curriculum quality (low-level test preparation festers in a high-stakes environment); moral (turning up the heat increases the amount of cheating); and simple fairness (how can schools divvy up credit among all the teachers who contribute to students’ success?).
So why are folks still talking about individual merit pay when it’s clear that it won’t work? Because the idea of holding teachers accountable for their students’ test scores sounds so obvious—and U.S. Secretary of Education Arne Duncan and a bunch of powerful politicians are enabling that gut feeling. States that didn’t include student achievement in end-of-year teacher evaluation and compensation worried that they stood very little chance of winning desperately needed Race to the Top funds.
Is there some way to make a silk purse from this sow’s ear? I believe there is. Let’s start with four points:
First, those who advocate performance-based accountability are absolutely right that student achievement needs to be front and center. It’s not enough to observe teachers’ classroom performance; we need evidence that students have learned.
Second, research has clearly established that teachers and principals make a huge difference to student achievement. They shouldn’t be ducking responsibility.
Third, when people are acknowledged for a job well done, it’s affirming and energizing. That’s true even for idealistic and intrinsically motivated educators.
And finally, everyone knows that the current teacher-evaluation process is broken; raising the stakes might mobilize us to fix it.
But how can schools focus on student learning and reward effective educators without running into the problems listed above? The key is making smart choices on (a) who gets rewarded when students do well, (b) what’s measured to determine rewards, and (c) the rewards themselves.
• Who gets rewarded. Should it be individual teachers? Teacher teams? Or the entire school staff? I believe the most productive choice is teacher teams. Rewarding teams promotes collaboration where it counts—among the three or four kindergarten or Algebra 1 teachers who teach the same content to different students. Rewarding teams avoids the problems of individual rewards (idea-hoarding and silo-dwelling) and large-group rewards (freeloading by lazy and ineffective colleagues). Team rewards encourage colleagues to push all students to high achievement—and create a dynamic in which peers hold each other accountable. As University of California, Los Angeles, professor James W. Stigler has written, team accountability in Japanese schools was a key factor in the steady improvement of teaching and learning there in recent decades. ("Needed: Fresh Thinking on Teacher Accountability," June 4, 2010.)
• What’s measured. Should it be end-of-year standardized-test scores? Value-added standardized-test scores? Student gains on in-school assessments? Or teachers’ classroom skills? I believe the best choice is a combination of individual classroom performance (based on the principal’s observations) and team student achievement gains (based on in-school assessments). The good news here is that there are new ways for principals to get a much better sense of what is going on in classrooms and we have increasingly precise tools for measuring student learning during the year: scales of reading proficiency, rubrics for scoring students’ writing, open-ended math questions that uncover students’ understanding, and carefully worded multiple-choice tests that give insights into students’ mistakes and misconceptions. In-school assessments may not be psychometrically perfect, but teachers trust them more than standardized tests, and the results are far more timely and helpful.
A pat on the back isn't enough, but merit pay (for individuals, teams, or the entire staff) increases the chances of shenanigans and gaming the system.
Step one would be for principals to make frequent brief and unannounced classroom visits, with face-to-face feedback after each one; this avoids the well-known problem of seeing only glamorized lessons that aren’t representative of what students are experiencing day to day—and catches teaching problems in real time. Step two would be teacher teams’ presenting the principal with evidence of all their students’ learning gains each spring.
• The reward. Should it be a pay bonus? Positive year-end evaluations? Or praise from the principal? I think the best choice is a team score as one element in teachers’ annual evaluations. A pat on the back isn’t enough, but merit pay (for individuals, teams, or the entire staff) increases the chances of shenanigans and gaming the system. Each teacher’s final evaluation should be based on (a) the principal’s assessment of his or her classroom performance based on multiple visits and conversations, and (b) a collective score for the team’s student learning gains that year. Team accountability would create a powerful incentive for teachers to work together to solve learning problems during the year—and get all their students over the bar.
There is a role for monetary incentives in three areas: career-ladder opportunities for the most highly rated teachers to take on extra responsibilities for extra pay; incentives for the most effective teachers to work in high-need schools and subject areas; and denial of step increases to teachers with mediocre ratings (while, of course, moving to dismiss teachers with unsatisfactory ratings).
Here’s what this proposal would look like for a hypothetical 2nd grade team.
In early September, teachers conduct a baseline assessment of their students, agree on the best metrics for measuring learning, and decide on goals (for example, 85 percent of students scoring proficient and above in writing by June). The principal looks over the plan and suggests a few changes before signing off. The team then designs curriculum units, teaches them (using varied methods and materials), checks for student understanding during lessons and with gradewide interim assessments, and has follow-up team discussions at least once a week about what’s working and what isn’t. The principal makes frequent classroom visits with feedback, drops in on team meetings, and constantly chats with teachers about methods and materials.
Toward the end of the year, the team collects the most recent data and meets with the principal to present its value-added report. There’s a discussion of accomplishments, problem areas, and curriculum revisions, and the principal gives the team an overall 4-3-2-1 score on learning gains, with commendations and suggestions.
Over the next few days, the principal meets individually with each teacher, asks for input, rates classroom proficiency and outside-classroom performance, adds in the team learning score, and gets the teacher’s signature.
This approach is far more powerful than individual merit pay. It puts the focus on student learning, harnesses timely student data to boost team collaboration, and rewards teams that get results. It immerses principals in the teaching and learning process, doesn’t overwhelm them with data (a principal in an average-size school would be looking at six team reports vs. 35 individual teacher reports), and makes year-end evaluations far more robust. And it includes teachers who are left out of test-based accountability—art, music, physical education, computer, library, and the primary grades.
All teachers in the building—and administrators, too—would be using evidence of student learning to continuously fine-tune their craft—a powerful engine for improving teaching and learning and making a dent in our persistent and troubling achievement gap.