Increasing the Content Focus of Teacher Evaluation

In this Harvard Educational Review article, Heather Hill (Harvard Graduate School of Education) and Pam Grossman (Stanford Graduate School of Education) argue that teacher evaluation needs to (a) use subject-specific observation instruments, (b) involve more content experts, and (c) design systems “in which feedback from observations is both accurate and usable in the service of improving instruction.” 

On the first two points, Hill and Grossman acknowledge that “content expertise is not always necessary: some elements of teaching – managing behavior, building a safe climate for learning, motivating student effort – are common across subjects, and, to the extent that a teacher requires assistance with these areas, a well-trained generalist observer might make a large difference in teaching and learning outcomes.” But there’s another layer of observation, they say, and current practices “ask us to believe that teaching kindergarten requires the same set of practices and knowledge needed to teach high-school algebra.” A principal could focus on classroom management and entirely miss important problems in the content.

On the third point, Hill and Grossman note the importance of improving the expertise of principals and other classroom supervisors and say, “One of the challenges for any observation instrument is getting the grain size right. By grain size, we mean the scope and level of detail around desired practices… Grain size matters in both the design and use of observation instruments. The more specific the grain size, the more specific the feedback for teachers can be.” But the more specific these are, the narrower the group of teachers that can be evaluated. 

Hill and Grossman also address the question of how many classroom visits and how many follow-up coaching sessions are needed to make significant improvements in classroom practices. “Research suggests that coaching programs that are successful in supporting improved student outcomes provide at least monthly coaching sessions,” they say. Two to four visits a year – the frequency suggested by many reformers – are clearly not enough to make much difference, but asking principals to make more visits to each classroom is seen by many as impractical. 

And there’s the further problem of grain size – giving teachers feedback that’s detailed enough to be helpful. This is why Hill and Grossman suggest more subject-specific evaluation instruments and more attention to the content training of principals and other supervisors. Follow-up with teachers is a perennial problem: many principals “may have trouble scheduling follow-up conversations,” say Hill and Grossman; “they may not return to a teacher’s classroom for months, at which time both may have forgotten important details surrounding the original feedback.” 

“If these new evaluation systems are to have a chance of improving the quality of teaching,” they conclude, “policy makers must resist the urge to simplify the inherently complex nature of teaching. This will require grappling with how best to focus on issues of content in efforts to evaluate and improve teaching, how to select and develop raters with expertise in both observation and professional development, and how to concentrate resources to provide the kind of high-quality coaching that can actually have an impact on practice.”

“Learning from Teacher Observations: Challenges and Opportunities Posed by New Teacher Evaluation Systems” by Heather Hill and Pam Grossman in Harvard Educational Review, Summer 2013 (Vol. 83, #2, p. 371-384), 

From the Marshall Memo #497