Fariña floats possible evaluations fix that would require UFT reversal

The city schools chief has floated an idea to simplify the complex new teacher evaluation system, but carrying it out would require a major concession from the teachers union.

When the union and city were negotiating teacher evaluations in recent years, one of the many sticking points was how many factors teachers should be rated on when their classes are observed.

The union wanted teachers to be scored on all 22 components of a teacher-effectiveness rubric, while the city pushed for just seven of the rubric components. Ultimately, the state intervened last year and insisted on 22 components.

Now, well into the city’s first year under the new system, many principals report feeling swamped by all their rating duties, and some teachers wonder how fairly they will be rated on all those measures.

Enter new Schools Chancellor Carmen Fariña, a former principal committed to lifting unnecessary burdens from school leaders. At a private meeting with administrators in January, she raised the idea of greatly reducing the number of rubric components that principals and other evaluators have to rate teachers on, according to several people at the meeting.

To get state approval for that change for next school year, the United Federation of Teachers would need to sign off on it. That would mark a significant reversal for the union — but perhaps a palatable trade-off as it seeks billions in back pay and raises in the ongoing contract negotiations with the city.

Meanwhile, the change would undoubtedly cheer school leaders who have struggled to observe each of their teachers multiple times this year and rate them on the nearly two dozen components as required by the new system.

“I’m way behind,” said William Frackelton, principal of Soundview Academy for Culture and Scholarship in the Bronx, who supports the 22-component rubric in theory. “But in practice, how manageable is it? It’s a beast.”

At a meeting in late January with district superintendents and school-support network leaders, Fariña spoke about the need to support overburdened principals, according to several attendees. She suggested one way to do that would be to pare down the 22 instructional components that principals must observe and rate.

“She said, ‘That’s too many, we need to get it down,’” said Alan Dichter, a network leader. He added that he took Fariña’s comment as an “intention,” not a firm commitment.

The component question has not gone away since that meeting. At a conference for new principals Saturday, a veteran principal leading a workshop on evaluations said there could be fewer components in the future, but that the city is still discussing the matter with the teachers union, according to a principal who attended the workshop.

“Something good is cooking,” said the attendee, who requested anonymity because she had not been authorized to discuss the private training.

The state education commissioner imposed the new evaluation system last summer after a long city-union tussle over the details. Under it, 60 percent of teachers’ ratings come from subjective measures, including observations by administrators.

To rate teachers’ performance, principals or other evaluators must use a rubric known as the Danielson Framework. The rubric is divided into four “domains” of teaching: planning; classroom environment, which includes managing student behavior; instruction; and professional duties, such as communicating with parents and keeping records. Those domains are then broken down into 22 narrower components, such as cultivating a respectful classroom culture and sparking rich class discussions.

In its written submission for the state arbitration hearing, the UFT argued that the full 22 components are “essential” to measure the complexity of teaching. What it didn’t say, but what many read into the UFT’s position paper, was that requiring all 22 components could protect low-rated teachers from consequences that include firing: More components mean more potential points a teacher could contest if given a poor rating.

The city education department argued that teachers could be fairly rated using just seven Danielson components. It pointed to research that shows complex rubrics can overwhelm evaluators, leading them to rate disparate components similarly. It also noted that the city had used seven components during an evaluation pilot program. It cited evidence that the pilot ratings were accurate and that 93 percent of school leaders in the program said the seven components provided enough data to make fair assessments.

State Education Commissioner John King sided with the union on the issue of components, ruling that the Danielson rubric was “validated and was designed to be used in its entirety.”

As a result, New York City principals must rate teachers annually on all 22 components, for which they can use both observations and other evidence, such as teacher-created lessons and tests.

Many principals and other administrators have struggled to observe each teacher the required number of times, document their ratings and evidence, and give teachers feedback.

Frackelton, the Bronx principal, and an assistant principal must observe and rate 30 teachers. He said some school leaders respond to that pressure by filling in “cookie-cutter” explanations of their ratings on multiple teachers’ forms. He said he avoids using such stock language only by working on the forms until 10 p.m. some nights and on Saturdays.

“It’s really a lot of work to do it well,” he said.

The schools in the city’s evaluation pilot program did not expect to jump from seven to 22 rubric components when the official system launched this year, said Thandi Center, New York City director for the New Teacher Center, which was one of the city’s lead partners in the pilot. She said many principals have complained the new system “isn’t doable,” and teachers have expressed concern about the “credibility” of their ratings.

“I just think it’s untenable to introduce 22 components and expect it to be done well consistently,” Center said.

Phil Weinberg, the education department’s new deputy chancellor for teaching and learning, acknowledged principals’ concerns about the evaluations in a letter last week. He offered them advice for “reducing evaluator burden” and announced a survey and “listening tour” next month where the city will collect feedback from principals about evaluations. He also urged principals struggling to rate all their teachers before the June deadline to contact their support networks “immediately.”

If the city and union were to agree on an evaluation change for next year, they would have to jointly submit a request to the state.

If they ask to rate teachers on fewer rubric components, they would need to prove that all four domains will still be assessed and that the “integrity of the rubric” is preserved, said Julia Rafal-Baer, executive director of the state education department’s Office of Teacher and Leader Effectiveness.

She noted that some districts have approved evaluation plans that guarantee all Danielson domains will be assessed, but not all 22 components will be rated. For example, Webster Central School District’s plan says any observed components can be rated, but only seven specific components absolutely must be rated.

Rafal-Baer added that it would be “interesting” if the city teachers union agreed to fewer components, since the UFT “really felt very strongly about having all 22 components” when it pitched its evaluation plan last year to the state.

The union is currently pushing for more than $3 billion in back pay in contract negotiations with the city, along with a pay hike for the future. Teacher evaluations are part of those negotiations, and the UFT could potentially use a component-number change as a bargaining chip.

A UFT spokesperson declined to comment, citing the union’s policy to avoid public negotiations.

A city Department of Education spokesman declined to comment on Fariña’s remarks or possible evaluation changes, saying the city’s focus is on “improving classroom instruction.”

“Through meaningful observations and feedback under the evaluation system, it’s our goal to help educators hone their craft,” said the spokesman, Devon Puglia.