Learn by doing — it’s a well-worn mantra, and sometimes it works. But after you’ve learned to do something, how do you learn to do it better? For teachers, who still work mostly in isolation, the risk of hitting a plateau — of doing, but not growing, or of getting stuck in a bad habit of practice — is high.
But “learning by doing” can work in a more focused way when the “doing” is guided by a successful peer and structured around a particular task. A new working paper just out from the National Bureau of Economic Research has demonstrated that for teachers (and perhaps workers in many other sectors), there is a tangible value in learning from colleagues.
Researchers from Brown University and the Harvard Graduate School of Education found that pairing highly skilled teachers with lower-performing colleagues at the same school, then asking them to work together for a year on specific skills, produced meaningful and lasting improvements in teacher skills and student performance.
The study, carried out with the Tennessee Department of Education, encompassed 14 elementary and middle schools and 136 teachers in Tennessee. Using data captured by that state’s intensive teacher evaluation protocol, researchers identified teachers who were weak in one (or more) of a constellation of assessed characteristics, and then matched them with teachers who were strong in corresponding areas. School principals reviewed the matches, revising them if needed, and then approached each pair and asked them to work together on improving instructional skills in the areas the low-performers needed to bolster.
Principals encouraged the pairs to look at each other’s evaluation results, observe each other’s teaching, talk about strategies for improvement, and follow up with one another throughout the school year.
The study found that the program — called the Instructional Partnership Initiative — has substantial effects. Researchers found that students of low-performing teachers who’d been randomly selected to join a partnership scored 12 points higher, on average, on standardized tests than students of low-performing teachers who didn’t join a partnership.
That gain is roughly equivalent to the difference between being assigned to an average teacher and a low-performing one, the researchers say. It is also at least as large as the difference in performance between a novice teacher and a 5- to 10-year veteran. And the researchers found that these improvements in teacher performance lasted and perhaps grew over the year following the experiment.
It’s easy to draw inferences about the importance of coworkers in our development on the job, but until this study, there hadn’t been much quantitative evidence of the role they play in our learning, the researchers say. They cite one earlier study that found that teachers’ performances improved when higher-performing teachers joined the faculty in the same grade. This new study is more direct in its measurement of the effects of a focused partnership between colleagues.
The findings — that teachers at all levels of experience can learn new skills from peers that translate into gains for students — might help fortify a new and less costly approach to professional development.
“The peer partnerships we study — a kind of one-on-one, personalized approach to on-the-job training — are designed to focus on practical, day-to-day problems,” says Harvard Graduate School of Education’s Eric Taylor, who co-authored the study with John Papay, John Tyler, and Mary Laski from Brown. “They offer as a solution to those problems the experience and advice of someone you know, someone who works under the same challenges.”