What's Next: Evaluation As Narrative

Stephen Sawchuk’s recent
Ed Week article on the new wave of teacher evaluation systems—and
the high ratings they're generating—has likely left many reformers scratching their heads.

Stephen points out that in states like
Tennessee and Florida, which are hotbeds for Michelle Rhee–type teaching
policies, the vast majority of teachers are quite effective. Even in Florida,
where teachers are judged by standardized test score gains, 97 percent received
top ratings. In Tennessee’s new system, more than 16 percent
of the teachers were rated with the lowest score on test score measures. Only 0.2
percent of those observed by principals were categorized as ineffective.

While
there are many policymakers—and virtually all of our nation’s 3.2 million
teachers themselves—who want to see more rigorous and valid evaluation systems
that strengthen teaching practices, there are also reformers who want nothing
more than to identify ineffective teachers and dismiss them. But overall, we need to shift our focus from what's not working for the ineffective minority to how we can ensure continued growth for the effective majority.

Although scholars have long pointed out the
problems of current teacher evaluation systems (see Art Wise and Linda
Darling-Hammond’s 1984
research (PDF) and its seminal findings), there is still much
to learn about the validity of the programs that are being put into place
today.

Even when pains are
taken to create a fair and reliable ratings system, such as the Measures
of Effective Teaching project, flaws still abound. Jesse
Rothstein and colleagues found
that in each of the multiple measures (standardized test score gains, classroom observations, and student engagement ratings) used in the carefully designed project reflects
a distinct dimension of teaching. They also noted that “none of the three types
of performance measures captures much of the variation in teachers’ impacts on
alternative, conceptually demanding tests.”

What none of these analyses suggest is how teaching conditions must change before we can design and implement a valid and reliable evaluation
system.

Both principals and teachers must
have more time to look carefully at
the data and make sense of it. In top-performing nations like Singapore, teachers
only teach students 18 to 20 hours a week. Here in the U.S., we devote little
organizational space to assembling sound evidence, analyzing it, and
applying professional judgment to determine who is effective and why—and how to
help them get better.

Singapore's core competencies for teachers, via the National Institute of Education, Singapore. (Click to enlarge.)

In Singapore there is no obsession
with standardized test scores. Teachers are evaluated on how they nurture the
child, work with parents, and care for both the heart and mind. In Singapore,
evaluation is not a number (or three of them); it is a narrative that encourages teachers to expand their teaching repertoire, select a
career track, and take those developmental actions that lead to greater competence
and higher levels on a career lattice. Teachers’ evaluations are based on the
experience level of the teacher, since the level of competence expected of a
new teacher is much lower than expectations for senior teachers. But most
importantly, the evaluations don't just assess current performance. They also
consider a teacher’s potential.

Will policymakers start
viewing evaluation differently? I suspect they will, if only they listened more
carefully to the millions of teachers nationwide who want an evaluation
system that will genuinely improve their practice and their students’ learning.