How to build a better teacher

The debate — and controversy — about the need to evaluate teachers and improve public education from pre-kindergarten to graduate school rages on every level from the local to the national, but the most basic question yet to be answered is this: How can we know what actually works in the classroom?

Evidence.

That's the conclusion reached during an academic symposium aimed at ways of bringing scholarly rigor to the art of teaching, held recently at the University of Virginia. The symposium, chosen by the U.Va. President Teresa A. Sullivan, focused on higher education teaching and sent an unambiguous message: It is time for every professor to pay heed to what, and how, their students are learning.

Higher education has recently been under attack for failing to measure what students actually learn. But how to evaluate the effectiveness of teaching? Few tools are available. It's not that teachers don't value the importance of teaching — it's just that evidence on what works has been usually limited to not-very-helpful test scores and student evaluations.

But a system designed at the U.Va. Curry School of Education, the Classroom Assessment Scoring System, or CLASS, is helping to change that. CLASS, which studies interactions between teachers and students and how those interactions improve learning, is providing the evidence we need to evaluate teacher effectiveness.

* * * * *

In November, U.Va. became part of a $40 million effort, the National Center on Quality Teaching and Learning, with the mission to improve teaching and learning in Head Start. The CLASS system is the cornerstone of the new Head Start accountability plan. It is the only observational teacher-assessment tool that captures teacher behaviors linked to student gains and that has been proven to work in tens of thousands of classrooms. It provides quantitative assessment of teaching quality.

The need for this kind of verifiable academic evidence in higher education was the focus of the daylong symposium at U.Va. Addressing the issue was the keynote speaker, Lee S. Shulman, president emeritus of the Carnegie Foundation for the Advancement of Teaching.

"Not only can professors apply the same scholarly rigor to teaching that they do to their research, they should be required to," he said. "If poor methodology is suicidal in a research paper, why isn't poor pedagogical methodology also suicidal in the review of a faculty member?"

And then the presentations began, each designed to identify ways in which the collection and analysis of evidence related to teaching has been used to improve learning. More than 150 faculty members from every academic unit of the university participated in 32 discussions on using evidence to improve teaching.

Hundreds more — including students — attended, reaching across the disciplines to share ideas about what works to increase learning. They heard about results and measurable data from developments in curriculum, community outreach, collaborative teams and the use of technology in engineering and health care education. Engineers, physicians, foreign language instructors, law professors and nursing professors all shared their ideas.

* * * * *

The symposium collected and shared recent research that can help teachers teach better and provided an invigorating look at new ways of teaching and transferring these techniques to other teachers through innovative technologies. And all of the conclusions of the presenters were based on information supported by careful study and results-based evidence — just what Shulman and I say are needed.

A few years ago, U.Va. established the Center for the Advanced Study of Teaching and Learning, or CASTL, to study the teaching process systematically and rigorously to solve specific problems. In one study, researchers worked with preschool instructional teams in five Virginia Head Start classrooms, providing one-on-one coaching sessions that included feedback based on classroom observations. They then gathered data about how well the teachers were able to apply what they learned.

Early results show that the teachers who participated in the sessions were much better equipped to handle problem behavior in the classroom than they were prior to intervention. They had learned preventive strategies, such as limiting the number of children in a center at a given time and making adjustments in transition routines. They also taught children who needed more assistance to use phrases during play so they could be better play partners. As a result, the number of instances of problem behavior decreased significantly compared to the prior year.

As this approach to teaching expands — with improvements in measures of teaching and of student learning and the development of proven classroom models — we have the power to transform education in America from pre-kindergarten through graduate school.