Photo by Sam Hodgson
Crawford High School calculus teachers Jonathan Winn and Carl Munn and their team created a grassroots data evaluation system that boosted teachers' performance. The school district all but ignored their efforts.
This story also ran in the May 2012 issue of San Diego Magazine.
School superintendents across America are talking tough. The time has come, they say, to get rid of failing teachers, or at the very least to identify them so that weaker teachers can get help to become more effective. No longer should students suffer the ignominy of an educator who isn’t interested, willing, or able to make them learn.
For decades, schools have relied on a principal passing through a classroom once a year or every few years to eyeball how a teacher is doing. Today districts across the country say there’s another way.
They’re using reams of test score data to watch the impact each teacher has on his or her students throughout the year, learning whether students gained or lost ground under each teacher.
And they’re adding that measurement to the teacher’s evaluation. They use it to find stars, to get help for struggling teachers and, in some cases, to dispatch failing teachers like they’ve never been able to before.
In New Haven, Connecticut, the school district pushed out about 2 percent of its teachers last year, after extensive evaluations revealed those educators either couldn’t or wouldn’t improve kids’ test scores. Those evaluations had been crafted hand-in-hand with the local teachers union, which embraced reform in exchange for increases in pay and benefits.
In Houston, former San Diego Unified School District superintendent Terry Grier has overseen a radical redesign of the teacher evaluation process. Grier says there’s no place for underachieving teachers in Houston’s schools, so educators who don’t improve have been shown the door.
And in Los Angeles, superintendent John Deasy has made redesigning teacher evaluation a cornerstone of his leadership.
Tackling California’s powerful teachers unions and navigating legislation that crimps his ability to dismiss bad teachers has been tough, Deasy acknowledged, but inaction’s not an option.
“This is both a moral and a legal imperative,” Deasy said.
The San Diego Unified School District, however, isn’t interested in this revolution.
|File photo by Sam Hodgson|
|Former San Diego Unified Superintendent Terry Grier is now the head of Houston Independent School District, where he’s implementing tough reforms on teacher evaluation.|
Today, teachers are evaluated in San Diego in much the same way they have been for decades.
Once every year or two, with advance notice, principals pay a perfunctory visit to each classroom. After a brief, formal observation, the principal completes a three-page evaluation form. Teachers are rated on the form as either Effective, Requiring Improvement, or Unsatisfactory.
The vast majority of local teachers receive an evaluation that says they’re effective, which isn’t surprising to many local principals.
“I mean, come on! If you can’t pull it off for a formal evaluation once every couple of years, for one lesson, then you really shouldn’t be a teacher,” said E. Jay Derwae, principal of Marvin Elementary School in Allied Gardens.
The cursory evaluation system in place at San Diego Unified was the norm across the United States until fairly recently. But as education reformers began to realize that a half-century of their efforts had done little or nothing to push up student achievement, attention began to focus on the sticky topic of teacher evaluation.
Districts across the country began jumping on the reform bandwagon. They’d been pushed there by the jarring success of the documentary “Waiting for ‘Superman,’” which reviled school districts for doing little to weed out underperforming teachers, and pressure from the Obama administration to revamp teacher assessment tools.
As the movement picked up speed, progressive superintendents began to coalesce around an evaluation process that had long been pushed by reformers: value-added metrics.
Very few people in the education community truly understand how value-added metrics work. That opacity is a main reason evaluation systems based on value-added metrics have proven so controversial.
The basic idea is to judge teachers not on how high their students’ test scores are at the end of the year, but on how much kids’ scores have improved while they’ve been sitting in the teacher’s classroom.
The method tracks the progress of each and every student, then compares that progress to how much each student was expected to improve at the beginning of the year. By analyzing how much the students in each classroom have improved, districts can identify which teachers are consistently pushing kids’ scores up, which are keeping scores flat, and which are failing to improve.
In places like Los Angeles and New York, those scores have been made public, based on the argument that parents deserve the information.
But just as its popularity has boomed, there’s been an equally forceful movement against value-added metrics.
Several once-bearish proponents of value-added metrics, including Linda Darling-Hammond, a former top education advisor to President Obama, now rail against the model. They argue that the margins of error are far too high to make such analysis meaningful in even the most complex of statistical models.
Test scores are just as likely to be raised or dropped by changes in a student’s socio-economic status or health, or by economic factors that affect classrooms, like swelling class sizes or dropping budgets for materials, as they are by a teacher’s ability, those critics argue.
There are other serious concerns, too: Critics worry that putting teachers on the hook for their students’ test results inevitably leads to “teaching to the test,” sterilizing classrooms into factories for rote learning.
These concerns have added a wild swirl of controversy that has only been intensified by high-profile media reports on value-added scores. Both the Los Angeles Times and The New York Times have caused uproars by publishing teachers’ value-added scores, stories that have inevitably been spun off in clichéd headlines about each city’s “Worst Teachers.”
None of that criticism has halted the march of data.
School districts from Washington, D.C. to Tennessee have plugged value-added metrics into their evaluation systems, filtering out teachers who aren’t pushing test scores up and, in some cases, firing them for consistently poor results.
In February, New York state legislators inked a deal with teachers’ unions that will phase in value-added scoring until it accounts for 20 percent of teachers’ evaluations.
And in Houston, superintendent Grier said 150 teachers were asked last year to take a buyout or get the sack, after evaluations based largely on value-added scores identified them as ineffective.
Grier and others say the method is just one facet of evaluation. That data is simply used to identify if a teacher’s performance warrants further examination, he said.
“It’s like if you have a really bad fever,” Grier said. “A fever is a symptom that something’s wrong, it’s not the problem itself. If a teacher has poor value-added scores, that’s a red flag. That’s when the principal needs to be going into that classroom and doing more observations.”
Why Be Divisive When You Can Be Productive?
|File photo by Sam Hodgson|
|District leaders such as school board trustee Richard Barrera say they’re encouraging principals to use data to help improve teacher performance, but principals across San Diego Unified are confused about what they’re supposed to be doing.|
The San Diego Unified school board and school superintendent Bill Kowba know all about value-added metrics.
They’re just not interested in using them.
District leaders shrug off value-added as a fad, saying it’s yet to be proven to push up student achievement. Similarly, they say revamping the district’s teacher evaluation process would be a sideshow that would detract from the serious work they’re doing to improve teacher performance.
Instead of imposing an aggressive, punitive evaluation process, San Diego Unified’s leaders said they’re building a “grassroots” system to help teachers get better at their jobs.
They say principals have been encouraged to use testing data proactively to identify struggling teachers and give them the support they need to push their students’ scores up.
“This shouldn’t be approached by the district throwing something at teachers that the teachers don’t believe in,” said trustee Richard Barrera. “For the conversation to be productive, it has to be something that teachers believe in, as well as the district.”
But while some district principals have embraced the use of data to help their teachers, they appear to be the exception, not the rule. And in at least one school, district leaders have all but ignored an effort by teachers to create exactly the type of proactive, data-driven system Barrera said he’s seeking.
There’s another big problem with the district’s gradual approach to identifying struggling teachers: It does nothing to tackle a formal evaluation system that many principals say is cumbersome and woefully inadequate.
Reforming that process will require an open dialogue between the district and its teachers union.
But the relationship between the district and the union has devolved to the point of being toxic. And even if they did begin to work together, the district is currently begging for teachers to take pay cuts to keep it afloat and avoid massive layoffs.
There’s likely little room for more concessions, even if district leaders wanted to change evaluations.
“This has been a non-starter in the district for as long as I can remember,” said Scott Himelstein, an academic and former secretary of education for the state of California.
A ‘Smokescreen’ For Reform
|File photo by Sam Hodgson|
|Principal Tavga Bustani created a data-driven process that greatly boosted teacher performance. She appears to be the exception, not the rule, at the district.|
To fully understand the power of the district’s approach to boosting teacher performance, Barrera said, you need to talk to a couple of principals who are doing amazing things with data.
He pointed at one school in particular: Edison Elementary in City Heights, whose principal, Tavga Bustani, has turned the once-struggling school around.
Bustani’s hard to get hold of.
Mornings are out. That’s because she spends several hours every morning in classrooms, observing teachers, assessing, analyzing, talking, and watching. She’s looking for gaps in her data, for signs that reinforce the stories told on the pie charts, line graphs, and scattergrams plastered on the walls of her office.
Bustani checks up on the data for her teachers every two weeks. Teachers who aren’t meeting test score and other assessment goals are given extra scaffolds and supports: They get extra training sessions, or receive more observation from Bustani. Sometimes, she pairs teachers up to learn from each other.
Using her data-driven system, Bustani’s getting excellent results: When she first came to Edison, only 21 percent of students were proficient in English language arts; now that figure’s 67 percent. In math, those figures have gone from 35 percent proficiency in 2008 to 74 percent today. Bustani’s not sure how many other principals are following her lead, but she has an idea.
“I think the level of rigor and the expectation of accountability that I have for my teachers and my students is probably not what is the norm,” Bustani said.
More than a dozen interviews with principals from around the district suggested she’s right.
Barrera and district chief of staff Bernie Rhinerson said principals have been told to revamp the way they use data to identify struggling teachers as part of a program called Professional Learning Communities.
That’s news to most of the principals interviewed for this story, who said the program is actually all about sharing teaching techniques, and has little or nothing to do with using data to identify teachers who need help.
“That’s nonsense,” said Esther Omogbehin, senior principal of Lincoln High School. “That’s not what Professional Learning Communities are about, and that’s not what we use them for.”
“This is a smokescreen, and they don’t even understand their own smokescreen,” Omogbehin added.
Barrera said some principals are further along in the effort than others, and expressed concern that some principals don’t envisage the Professional Learning Communities in the same way he does.
“We will have to talk to those principals,” he said.
At the same time the district’s having problems rolling out its vision for improving teacher performance using data, it’s also proven deaf to at least one school that’s done exactly what Barrera says the district wants.
‘Data Tells A Story’
|Photo by Sam Hodgson|
|The team at Crawford High is now threatened by layoffs and reorganization of the school.|
Calculus wiz Jon Winn points at a column on his laptop.
“See that, there! That was a teacher who was struggling.”
He clicks to another slide on his Excel spreadsheet.
“And look, here’s a teacher who’s killing it. Look at his scores!”
Winn is one of the San Diego Unified School District’s wunderkinds. Last year, he was both named a district teacher of the year and won a coveted national award for outstanding teaching.
When he’s not dressing up as a member of a marching band to teach kids calculus, Winn has a pet project: In 2010, he and fellow math teacher Carl Munn started tracking reams of data from students’ test results at their school, Crawford High.
The two data wonks plugged thousands of figures into spreadsheets and began analyzing it, looking for patterns.
“Data tells a story,” Winn likes to say.
By figuring out how many students in each class were consistently performing well, and by comparing one class to another, Winn and Munn began identifying which teachers on the campus needed help.
One of those teachers was Ken Herschman, whose students consistently scored lower in algebra than kids in other classes.
“Ken comes into our meeting and says, ‘I’m throwing in the towel. You guys just tell me what to do, because I can’t figure it out. I cannot figure out, on my own, how to reach these kids,’” Winn said.
The team members at Crawford galvanized to help their colleague. For a year, Herschman gave up his daily preparation periods, opting instead to spend the time being mentored by Munn. He sat in on Munn’s classes, observing the master teacher’s techniques, which he replicated in his classroom.
Herschman’s students’ algebra scores soared. In a year, he went from being ranked 27th in the district to fourth.
“It just makes you look forward to coming to work,” Herschman said. “You really feel like you’re impacting student learning and student achievement. It’s a much more positive experience.”
Winn dubbed the algebra teacher’s journey “The Herschman Model.”
Without even knowing it, Winn and the Crawford team had created exactly the sort of collaborative grassroots effort district leaders say they’re looking for.
Thrilled by the results of their experiment, Winn and his colleagues wrote a 10-page report on what they had found.
Last year, they sent it to superintendent Bill Kowba.
“He loved it. He sent it out to all his deputies and everything and said ‘Let’s get this ball rolling,’” Winn said.
What happened next?
Winn waited and waited for a champion in the district’s higher echelons to develop or expand the model he and his colleagues had pioneered.
“Honestly, the ball was dropped. Nobody took it on,” Winn said.
A few months later, the four campuses at Crawford High were recommended for consolidation back into a standard high school.
“They said we didn’t have a plan for student achievement,” Winn said. “I feel like I’m going to scream. That’s what causes teachers like me to get 100 kids down at the board of education stomping and shouting, ‘You’re not listening.’”
This year, Winn was one of more than 1,600 district teachers to receive a notice he might be laid off.
A Problem Of Trust
Push San Diego Unified’s leaders hard enough and they’ll admit something: Even with their efforts to boost teacher performance, they still need to revamp the district’s formal teacher evaluation process at some point, too.
But doing that is far from simple.
Districts like Washington, D.C. and Houston haven’t just changed the way teachers are assessed; they’ve also upped the consequences of getting a negative assessment. Basically, they’ve made it possible for administrators to fire teachers based on their evaluations.
In San Diego Unified, as in many other districts, the process for dismissing a teacher is convoluted and complicated. It’s been built up by years of labor negotiations, so teachers now enjoy protections that require principals to prove over at least a year that the teacher is unsatisfactory at their job before they can be fired for poor performance.
Unraveling those protections would require a lot more than simply allowing or even encouraging principals to write up more rigorous teacher evaluations. And it could only be accomplished through negotiations with San Diego’s powerful teachers union.
That’s not a fight the school board wants to have right now.
The district’s currently too ensnared in squeezing out other concessions from its labor unions to also ask them to consider redrawing the teacher evaluation process.
“It’s just not a priority,” Rhinerson said.
The real problem here, ultimately, is one of trust.
The school district doesn’t trust the teachers union, which has become increasingly isolated and confrontational in the last few years, to work with it to create a new evaluation system. And the union shows no sign of sitting down with the district to discuss any issue, least of all one as contentious as teacher evaluation.
Until that conversation begins, San Diego Unified will remain exactly where it is today: Outside the teacher evaluation revolution, watching other districts embrace new assessment techniques that they can only attempt to replicate with the concurrence of their principals, teachers, and unions.
Will Carless is an investigative reporter at voiceofsandiego.org currently focused on local education. You can reach him at email@example.com or 619.550.5670.
Like VOSD on Facebook.
Value investigative reporting? Support it. Donate Now.