State officials admit problems with new teacher ratings

Teachers in Long Island classrooms, such as this

Teachers in Long Island classrooms, such as this one in a May 1, 2013 photo, are evaluated by a system using student test scores, classroom observations and factors such as contacts with parents. (Credit: Heather Walsh)

State education officials for the first time are acknowledging significant flaws in the accuracy of new teacher job ratings, forcing them to push back the deadline for public release of data showing how teachers rate in local districts until at least mid-March.

Department of Education staffers still are refining teacher evaluations for the 2012-13 school year, submitted by districts in October, spokesman Tom Dunn said last week. The job ratings measure teacher performance through a combination of students' scores on tests, classroom observations and additional factors, such as contacts with parents.

"We are taking every measure possible to ensure that our data release is accurate and protects personally identifiable information," Dunn said.


DATA: Opt-out numbers by district
LI test scores - ENGLISH: Grade 3 | 4 | 5 | 6 | 7 | 8
LI test scores - MATH: Grade 3 | 4 | 5 | 6 | 7 | 8
MORE: BOCES proposes changes | Take a sample math test


"There have been problems," he added. He did not provide any examples or say how numerous the flaws are.

"All the districts in the state -- nearly 700, excepting New York City -- have submitted the data," Dunn said. "It's a significant set of data, and it's never been reported before. We want to get it right the first time."

 

Fine-tuning the fixes

The work is aimed both at correcting any glitches in the data and at ensuring eventual release of numerical counts would not result in any teachers being personally identified, Dunn indicated. Dunn said the data would be released "sometime this winter."

The teacher evaluations could take on greater importance in the months ahead. Gov. Andrew M. Cuomo, in his budget, has proposed earmarking $20 million in state money for merit raises of up to $20,000 each for teachers earning the highest evaluation mark.

The 2010 evaluation law, which has been a major source of controversy, for the first time linked teachers' job ratings to their students' performance. Educators earn one of four job-performance rankings -- highly effective, effective, developing or ineffective. Teachers who are rated "ineffective" for two consecutive years can lose their jobs. Districts and teacher unions had to agree on local evaluation plans.

"Everybody knew the rules of the road, which produced ratings for teachers," said a member of a Cuomo-appointed commission that recommended state financial incentives for outstanding teachers, who spoke on condition that he not be named. "What the governor's system aims to do is reward teachers who were rated 'highly effective.' "

The evaluation system is complex and divided into three components, with 20 percent of ratings based on state test results when available, 20 percent on standardized tests selected by local districts, and 60 percent on classroom observations and other subjective criteria, as agreed upon between districts and teacher unions.

The law requires annual public release of results in all school districts, including 124 on Long Island.

Job ratings for individual teachers can be released only to students' parents -- a provision built into the statute to protect teachers' privacy.

The department has said it is working to ensure that any release does not violate the privacy provision -- for example, releasing job ratings in districts with small numbers of teachers could mean the identification, however inadvertent, of an individual teacher or teachers.

The state's portion of teacher job ratings for the 2012-13 school year were given to individual districts in August. Districts sent their part of the evaluations back to the state in October.

On Oct. 22, the agency released a statewide composite of the data, which showed 49.7 percent of teachers were rated "highly effective," 41.8 percent "effective," 4.4 percent "developing" and only 1 percent "ineffective."

 

Keeping the data private

The agency never has publicly released districts' job-ratings data for the 2011-12 or 2012-13 school years. The first year the law was in effect it applied only to teachers at certain grade levels. Agency officials, including Education Commissioner John B. King Jr., have repeatedly promised over the past 13 months to release the ratings.

The evaluation system -- which covers more than 126,000 teachers statewide, including more than 35,000 on the Island -- was an initiative pushed both by Cuomo and his predecessor, former Gov. David A. Paterson. The state's agreement to adopt a system that is tied in part to students' test scores was a major factor in New York winning nearly $700 million in federal Race to the Top money.

New York State United Teachers, a statewide union umbrella group, has called for a two-year moratorium in using test scores as part of evaluations. Carl Korn, a union representative, said the system has been "corrupted by the state Education Department's failed implementation plan."

Anecdotal accounts provided by school administrators, teachers and others across the Island in recent months have suggested multiple reasons ratings may not accurately reflect teachers' classroom performance.

Islip school administrators reported in the fall, for example, that eighth-grade math teachers there had received less-than-effective ratings from the state, even though 89 percent of their students passed Regents algebra exams that are at a ninth-grade level. The low ratings stemmed from scores on another set of state tests geared specifically for eighth grade.

State authorities have obtained a waiver from the U.S. Department of Education to prevent such "double-testing" issues in the future.

East Rockaway, meanwhile, is seeking state permission to amend its evaluation plan to deal with perceived glitches. Lisa Ruiz, who took over as the district's superintendent in September, said she discovered that some teachers who had received "effective" ratings in all three subcomponents of the evaluation system nonetheless got lower "developing" scores overall. "We certainly saw anomalies in our teacher scores, the first time around," Ruiz said.

East Rockaway is not alone. The State School Boards Association reported last month that nearly 40 percent of districts statewide had decided to revise plans for rating teachers and principals -- in some cases to develop more accurate measures of student achievement, and thus, of teachers' success in the classroom.

Some "growing pains" were to be expected in an initiative as complex and extensive as the state's evaluation system, association spokesman David Albert said.

advertisement | advertise on newsday

Vote

Do you think state tests accurately measure students’ progress?

Yes No

advertisement | advertise on newsday