Calculating influence and accountability

“A teacher affects eternity; he can never tell where his influence stops.” –Henry Adams

 “A policy maker affects children while the policy is in fashion. Common sense never goes out of fashion.” –Dan Brown

 First of all, it’s insane to me that an invisible legion of bad teachers are being scapegoated for the struggling of America’s public schools. Overwhelmingly teachers are already doing the best they can with what they’ve got, and there isn’t a shadow population of several million geniuses waiting in the wings to take their jobs and their salaries. Greater teacher accountability should not be the centerpiece of national education reform. But it is something worth exploring and improving.

It’s upsetting that in 2011, we seem to need a teacher’s influence quantified into a data point, printable in the newspaper. How do we calculate that influence?

Currently, high-stakes test scores seem easiest to tabulate and to wield for judgment. But what about the 70 to 85 percent of all teachers whose students don’t currently take high-stakes tests? Stephen Sawchuk’s article in EdWeek shows a range of stakeholders casting about as to how exactly to measure concretely the effectiveness of the gym teacher, the social studies prep teacher, the literacy coach, the pre-K teacher…

Here in the golden age of Data, we require algorithms for everything. At any point, if the data turns sour, sorry teacher, you don’t deserve a job anymore. In fact, you’re a villain for letting kids down. It’s impossible to envision this rush to generate new data models for high-stakes hiring-and-firing decisions being accepted in any other professional sector.

And yet, it’s being shoved down educators’ throats. Computers will tell us who is an effective teacher and who is an ineffective one.

Before we drive the train all the way off the cliff, my questions are these:

For teachers of students who do take the tests: Who is going to be inclined to work with lower-skilled children, the ones less likely to produce growth on stress-laden standardized tests? In Washington, D.C. last school year, less than 5 percent of the teachers who scored “highly effective” work in high-poverty Ward 8. More than quadruple that number of “highly effective”- rated teachers work in affluent Ward 3. Are the teachers worse in the poorer neighborhoods— or is it a lot harder to get test score gains in those schools?

For teachers of students who do not take the tests: How many good teachers will be fired because they don’t deliver whatever statistic is most valued in the newly created data model for their position? Campbell’s Law is: “The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor.” In other words, excessive teaching to the test destroys the reliability of the test. For non-testing teachers, whatever stat their career depends on will likely be delivered, but it will be at the expense of a richer educational experience for the students.

My solutions aren’t rocket science. It doesn’t require the complex opacity of credit default swaps to know if a teacher is doing a good job. It should be apparent based on: observing the teacher with students, observing student assignments (for richness, rigor, engagement, and growth), feedback from students and parents, and a continuing discourse with the teacher.

In March 2008, I blogged about a four-pillar proposal for a new accountability system in New York City. In response to the reductive and distorted single-letter grades handed out to schools by the DOE, the UFT developed a two-way accountability system judging schools based on:

(1) academic achievement;

(2) safety, order, and discipline;

(3) teamwork for student achievement;

(4) Department of Education accountability to the school.

Sadly, the proposal went nowhere and has now been deleted from the UFT website. But the fourth pillar— the one rarest seen and least discussed in public forums— is vital to making schools work. Teachers can’t be at their most effective in under-resourced or hostile environments. Their students can’t learn a lot with ill-chosen corporate curricula.

Many models are out there for schools that do great work. Reinventing the wheel with new accountability systems that deify data and testing take us down the wrong road. Remember Henry Adams.

Related categories: