July 1, 2015
As the national debate about how to improve teacher quality continues, new research led by the University of Michigan shows that teacher involvement in quality collaborations with colleagues can lead to improved teacher and student performance.
The study by Matthew Ronfeldt, assistant professor of education at the U-M School of Education, and colleagues from U-M and Vanderbilt University represents the first district-level view of collaborations and their relationship to achievement. The study not only looks at what kind of collaborations teachers find helpful, but it determines which ones most lead to student success.
"There is growing consensus that differences in teacher quality exist and that these differences matter for student achievement," Ronfeldt said. "Many policy efforts today assume quality is fixed or inborn, and focus on increasing the supply of 'best and brightest' into the profession to replace 'bad apples.' However, growing research evidence suggests that a teacher's quality is not fixed and depends a great deal upon school working environment and climate, and the quality of colleagues around her."
The researchers used surveys and administrative data to study more than 9,000 teachers in 336 schools to determine the amount and quality of collaborative efforts among teachers in the Miami-Dade County Public School System, the 4th-largest district in the country.
They found that 85 percent of teachers in the district were part of instructional teams, and that participation improved both the teachers' perceptions of the value of teams, and the quality of instruction. Almost 90 percent of teachers reported that collaboration in instructional teams was helpful or very helpful.
There were three general areas of collaboration: instructional strategies and curriculum; instructional approaches to groups or specific students, regarding classroom work, student discipline and class management; and approaches to assessment, including review of state test results.
Previous research has tied collaboration among teachers with achievement but those studies focused on the amount or extensiveness of collaboration, lumping all of it together regardless of its content or quality.
"Ours is the first study, to our knowledge, to develop measures for the quality of collaboration about different instructional domains and attempt to disentangle the relative impacts of collaboration around different topics," Ronfeldt said.
For example, the team found that high-quality collaboration about assessment, rather than about students or instructional strategies/curriculum, were better predictors of student math achievement gains. However, better quality collaboration across a range of instructional domains, rather than about a single domain, was most predictive of better student achievement.
"These results have important implications for school leaders looking for ways to boost student outcomes," said Jason Grissom, associate professor of public policy and education at Vanderbilt University. "Focusing on building teacher teams and providing meaningful ways for teachers to work together on the tough challenges they encounter can lead to substantively important achievement gains."
Previous research showing a positive relationship between collaboration and achievement did not rule out the possibility that better quality teachers might simply sort into schools with better collaboration. If so, then schools with stronger collaboration might happen to have better achievement gains without this relationship being causal.
"We use an innovative modeling strategy to show that a teacher's rate of improvement is greater in schools with better quality collaboration, as compared to the same teacher's rate of improvement in schools with worse quality collaboration, Ronfeldt said. "Differences in teacher quality are unlikely to explain these results because we are examining the same teacher in different settings.
"Though our analyses are correlational and do not permit drawing causal conclusions, our innovative modeling approaches and various tests of threats to validity provide some of the best pseudo-causal evidence to date that better teacher collaboration improves teacher effectiveness at raising student achievement.
"Our results indeed suggest that promoting better quality collaboration among teachers is a promising way to grow social and human capital in schools."
In addition to Ronfeldt and Grissom, other authors include U-M's Susanna Owens Farmer and Kiel McQueen.