Getting Serious About Measuring Collaborative Teacher Practice
Our guest author today is Nathan D. Jones, an assistant professor of special education at Boston University. His research focuses on teacher quality, teacher development, and school improvement. Dr. Jones previously worked as a middle school special education teacher in the Mississippi Delta. In this column, he introduces a new Albert Shanker Institute publication, which was written with colleagues Elizabeth Bettini and Mary Brownell.
The current policy landscape presents a dilemma. Teacher evaluation has dominated recent state and local reform efforts, resulting in broad changes in teacher evaluation systems nationwide. The reforms have spawned countless research studies on whether emerging evaluation systems use measures that are reliable and valid, whether they result in changes in how teachers are rated, what happens to teachers who receive particularly high or low ratings, and whether the net results of these changes have had an effect on student learning.
At the same time, there has been increasing enthusiasm about the promise of teacher collaboration (see here and here), spurred in part by new empirical evidence linking teacher collaboration to student outcomes (see Goddard et al., 2007; Ronfeldt, 2015; Sun, Grissom, & Loeb, 2016). When teachers work together, such as when they jointly analyze student achievement data (Gallimore et al., 2009; Saunders, Gollenberg, & Gallimore, 2009) or when high-performing teachers are matched with low-performing peers (Papay, Taylor, Tyler, & Laski, 2016), students have shown substantially better growth on standardized tests.
This new work adds to a long line of descriptive research on the importance of colleagues and other social aspects of the school organization. Research has documented that informal relationships with colleagues play an important role in promoting positive teacher outcomes, such as planned and actual retention decisions (e.g., Bryk & Schneider, 2002; Pogodzisnki, Youngs, & Frank, 2013; Youngs, Pogodzinski, Grogan, & Perrone, 2015). Further, a number of initiatives aimed at improving teacher learning – e.g., professional learning communities (Giles & Hargreaves, 2006) and lesson study (Lewis, Perry, & Murrata, 2006) – rely on teachers planning instruction collaboratively.
As school systems devote tremendous resources to examining the effectiveness of individual teachers, how can we encourage schools to make room for collaborative practices? How can we ensure that emerging teacher evaluation systems do not, in the words of Susan Moore Johnson, “reinforce the walls of the egg-crate school?”
Part of the solution might be rhetorical. As the research on collaboration referenced above suggests, teacher collaboration may have an important role to play in improving student achievement. There are likely to be returns to individual teachers (and their students) when they engage in productive interactions with their colleagues. Collaborative school structures can promote teacher learning and, by extension, teacher quality. Similarly, evaluation systems may make teacher collaboration more impactful. For example, although the initial teacher matching experiment conducted by Papay et al. (linked above) did not have a large enough sample to test for potential causal mechanisms, it stands to reason that one thing that may have driven improvement was that performance data from classroom observations provided an instructional focus for teacher pairs.
A second way to resolve these two policy perspectives is to expand the array of tools used to measure teaching quality formally. Schools naturally orient themselves around what is measured and what gets included in accountability systems; thus, the emphasis that evaluation systems give to classroom observations and student growth measures may at least be partially attributable to the fact that these measures were extant when the evaluation systems were developed. If we want schools to promote collaboration, then we need to know more about how effective collaboration might be assessed.
In our new Shanker Institute paper, “Competing Strands of Educational Reform Policy: Can Collaborative School Reform and Teacher Evaluation Reform Be Reconciled?,” my colleagues and I review the educational literature to identify how teams of educators working collaboratively might validly be evaluated. Finding little existing research, we then turn to other fields and professions for guidance on how such measures might be developed.
While educational research still has little to offer in deciding how to measure team performance within schools systematically, we found that much can be learned from other fields, including the military, the medical profession, and business, each of which has devoted considerable effort to improving the assessment of shared practice. In fact, since completing our paper, even more work on the topic of team performance has been published. The March 2016 issue of Medical Teacher includes findings from a study of the long-term impacts of team performance during medical school.
In reviewing this literature, two promising constructs stand out. Measures of team planning and team knowledge could gradually be integrated within evaluation and accountability systems to support professional learning and ensure that collaborative practices are rewarded. In short, our paper suggests that team performance can be measured, and measured rigorously.
New evaluation systems hold potential to improve the quality of teaching and learning in schools. But the likelihood that these policies will take hold and resonate with teachers will depend on whether they map onto the realities of teachers’ work. Many evaluation systems assume independence, but like other forms of work (e.g., surgery, the military), the teaching profession is one marked by shared practice. Teaching is situated within social organizations, and despite evidence that qualities of the school organization have an impact on teaching effectiveness, our profession lags far behind in how seriously we take collaboration.