Print

Teacher Evaluations: January's Dream Agreement Now On The Rocks

Misunderstanding On Key Issue Of Student Test Scores

By KATHLEEN MEGAN

May 23, 2012

In what appeared to be a diplomatic feat back in January, the state's Performance Evaluation Advisory Council heralded an agreement on a new teacher evaluation system.

Teachers union leaders, administrators, board of education representatives and state officials all endorsed a framework for evaluations that relied heavily on student peformance.

But now it appears that council members had conflicting ideas about what they had agreed to, specifically with regard to the percentage of a teacher's evaluation that would be based on students' performance on tests.

The disagreement surfaced recently as council members — and four working groups of that council — hammer out guidelines to implement the evaluation system. The council is expected to deliver recommended guidelines to the state Board of Education which must decide on them by July 1.

"We have a very fundamental disagreement here," said Sharon Palmer, president of the American Federation of Teachers Connecticut. "You see this is what happens when you rush. … When you are trying to fly the plane and build it at the same time … you move too quickly, and you end up with controversies."

More than 30 school districts have applied to become part of a pilot program, in which eight to 10 school districts will voluntarily test the evaluation system, said state Education Commissioner Stefan Pryor. The pilot program will start this fall.

The deadline for applications is Friday, and districts selected for the pilot will be notified by June 1.

The number of applicants "is a terrific sign," Pryor said. "I think it's a vote of confidence in the process."

At issue in the dispute among advisory council members is how much weight students' test scores should have in a teacher's evaluation. The advisory council's framework — which the state Board of Education approved in February — says that 45 percent of the evaluation will be based on "multiple student learning indicators," with half that measure based on standardized test scores.

The disagreement is on the other 22.5 percent. Teachers union leaders contend that this half was to include other measures of student performance — student portfolios or projects or other indicators — but not tests. Those representing school administrators and boards of education say they were certain that other half could include tests.

"There appears to be a disagreement as to what we agreed to," said Joseph Cirasuolo, executive director of the Connecticut Association of Public School Superintendents and a council member. "This isn't just details …"

Patrice McCarthy, a council member and deputy executive director for the Connecticut Association of Boards of Education, said she thinks teachers and administrators should have the flexibility to include other kinds of test scores in that second 22.5 percent.

Consider a teacher in an Advanced Placement calculus class, McCarthy said. "If one of the teacher's goals is to have a certain percentage of students get a 3 or above on an AP exam, shouldn't those results be a part of that teacher's evaluation?"

Cirasuolo said he is sure that a majority of the council members would never have agreed to limiting the role of test scores in a teacher evaluation to less than one-fourth.

Palmer, who is also a member of the advisory council, said she thinks that allowing test scores to account for slightly less than a quarter of teacher's evaluation is an appropriate percentage.

"We were really quite certain that was the understanding, but apparently there's some confusion," Palmer said.

She said closing the achievement gap depends on many factors that she believes are just as important as testing. "What are we doing to improve the attendance rate? What are we doing to reduce discipline issues? What are we doing to help stabilize the student population in the school? All of these things are terribly important for student performance."

Education Commissioner Stefan Pryor declined to specify his interpretation of the agreement, saying he hoped to "remain an impartial arbiter to some degree."

Pryor said "a good dialogue" has begun on the issue. "I believe we can reach resolution on this issue."

He also said the planned pilot program could help resolve the disagreement.

At a meeting Wednesday of one of the advisory council's working groups, Lori Rossomando, president of the Stamford Education Association, said that, with almost one-quarter of an evaluation based on tests, it's important that the other 22.5 percent not be.

Rossomando said tests "don't in any way measure a total child… We shouldn't want teachers looking at students in such a one-dimensional way."

Although no votes were taken, the working group seemed to agree that teachers should meet with an evaluator to set at least two but no more than four goals for the school year, and that the goals should be mutually agreed upon by teacher and evaluator.

Rossomando raised a concern: If teachers are going to be rated according to whether they meet a particular goal, she said, they may be less likely to pick a challenging goal.

David Calchera, director of public policy for the Connecticut Association of Public School Superintendents, said: "These are valid concerns and we've heard them for a long time. I thought they were taken care of, and, if they are not, we need to revisit them."