Good evaluation

A recent study finds that when teachers receive time and tools to collaborate with their peers, they are “more likely to teach effectively and to remain in the high-needs schools that need them the most.”

Getting effective teachers means making teachers effective.

Yes, that sentence came out as intended. Bear with me. The problem with trying to identify effective teachers — whether by student test score or principal observation or both — is that there is nothing inherent in either plan to make teachers effective. Sorting and ranking your teachers, which is what New York City’s current test-based value-added assessment is about, doesn’t improve the profession, unless you buy the version of school improvement that says that getting rid of instructional “underperformers” is all that’s needed.

Such sorting and ranking is the brainchild of widget-minded executives, and it won’t turn around a school. The fiction is that you rid your schools of “dead wood.” But it’s just as likely that you’ll rid your schools of new teachers, unconventional teachers and unlucky teachers, while maybe missing the one who should go.

Few educators or researchers endorse value-added assessments as the sole tool for judging teachers. The new teacher evaluation system hammered out by the UFT, NYSUT and the State Education Department last spring calls for basing 20 percent on student growth on state exams, 20 percent on locally selected measures of student achievement (to be negotiated with the local union) and 60 percent on such measures as observation and peer review.
But even that framework has its problems. Who’s doing the observing? And what are they looking for?

Evaluation as it used to be

Not long ago, our evaluations were worse. When I applied for the (now defunct) substitute social studies teacher’s license in 1992, I was sent to the old Board of Examiners where a man wordlessly looked at my paperwork and asked one question: What would I teach about India? Well, um, colonialism, I mumbled, and I guess also Gandhi. That was it. A few weeks later my license arrived in the mail.

Granted, that was an interview, not an evaluation, but it was the only evaluation I had before I was put before a room full of students.

So we probably don’t want to go back to the good old days since they weren’t all that good. What we want is an evaluation system that succeeds in enhancing the education of the children we teach. That is not going to happen via a sort-and-fire scheme of any kind.

But turn the whole enterprise 180 degrees and you get a different mindset. What are the conditions that build teaching skills?

Recently researchers have determined that standard teacher credentials — education and experience — are not the only predictors of student achievement. They see other “teacher quality” measures they struggle to identify. Is it enthusiasm, classroom management skill, verbal ability? The research is ongoing and fascinating.

But they are looking at the individual teacher level, and what this misses are the school conditions. Those may be the most important teacher quality factor of all.

Good schools make good teachers

“Teacher effectiveness has less to do with individual attributes and far more to do with the extent to which teachers work with each other and provide collective leadership for their schools and communities,” write Barnett Berry, Alesha Daughtrey and Alan Wieder in a 2009 paper, “Collaboration: Closing the Effective Teaching Gap,” from the Center for Teaching Quality. “Working conditions seem to matter a great deal for teacher effectiveness.”

Which conditions? Based on a national survey, they identify four: ample common planning time (“collaboration rarely ‘just happens’”); vertical collaboration across multiple grades as well as in a single grade; a structured agenda for looking at practice and student work; and an atmosphere of mutual trust.

“When teachers are given time and tools to collaborate with their peers, they are more likely to teach effectively and more likely to remain in the high-needs schools that need them the most,” the Center for Teaching Quality authors find.

Another recent study cited in the July 2010 paper “Professional Development in the United States” by Stanford University researchers and the National Staff Development Council found that in high-poverty schools teaming teachers led to vastly better student outcomes than in matched comparison schools in the same district.

Yet such teaching conditions are actually becoming rarer. “Teachers were only half as likely to report collaborative efforts in their schools in 2008 (at 16 percent) than in 2000 (when 34 percent did so),” the Stanford authors write. They also found that American teachers have much less time in their day to collaborate than do teachers in higher-achieving countries.

The new Race to the Top evaluation formula gives great weight to classroom observations where an administrator watches and rates teachers in action, using a rubric of agreed-upon good teaching practices. This step forward acknowledges that good teaching is about more than test drills. It may also give teachers some feedback on what, specifically, they need to improve.

But the next big step is to build collaborative practice into the daily routine of schools so teachers can learn from and build upon each other’s work.

There are some ineffective teachers, but as Berry and her colleagues say, “(m)any of those ineffective teachers were never sufficiently prepared or supported to succeed in high-needs classrooms — and simply removing poor performers will not ensure that effective teachers will be waiting in the wings to replace them. Specific strategies to spread the expertise of the most accomplished teachers may be the key to turning around low-performing schools.”