clock menu more-arrow no yes

Filed under:

Numbers show teacher evaluation system broken

Download pdf version of this story
* Nearly 100 percent of teachers in Colorado’s largest school districts received satisfactory ratings in each of the past three years, an indication the state’s system to improve classroom instruction is broken.

Education News Colorado requested teacher evaluation data from the six largest districts, all in the metro area, which serve more than 40 percent of public school students statewide.

The analysis found little difference between the results of evaluations given in affluent, high-performing Douglas County and those doled out in urban Denver Public Schools, where large numbers of students perform below average on state exams.

Fewer than 2 percent of teachers in either district – or in Adams Five-Star School District or in Jefferson County Public Schools – were told they needed to improve their instructional skills.

Two districts – another high-performer, the Cherry Creek School District, and the struggling Aurora Public Schools – declined to provide any numbers in response to requests under the Colorado Open Records Act.

“Unfortunately, we do not have existing records on the ‘the number of teachers who received unsatisfactory evaluations as determined by the teacher evaluation system,’ “ Aurora spokeswoman Georgia Duran said in the district’s response.

When Ed News pointed out the district’s evaluation form does include an unsatisfactory rating, Duran responded the district does track the data but doesn’t have “a ready means” of reporting it. The district pledged Tuesday to compile the data and release it within two weeks.

Districts won’t be able to hold back their numbers much longer. U.S. Secretary of Education Arne Duncan wants the data made public in return for the more than $100 billion that the federal government is investing in education under the American Recovery and Reinvestment Act.

“In exchange for this unprecedented funding boost, we are asking you to collect, publish, analyze and act on some basic information … information that will reveal both strengths and underlying challenges,” Duncan said in an April 1 letter to the nation’s governors.

He cited specifics including, “the number and percent of teachers … rated at each performance level in each local education agency’s teacher evaluation system.”

Colorado legislators, eager to secure as much of the federal money as possible, copied that language into a bill that was signed into law last week by Gov. Bill Ritter.

The data obtained by Ed News and interviews with administrators and teachers reveal little faith in the evaluation system created by state law “to serve as the basis for the improvement of instruction.”

“I will fully tell you I am so frustrated with this whole thing,” Jefferson County Schools Superintendent Cindy Stevenson said, days after a judge recommended a teacher be reinstated over district objections.

The case was one of two since 2002 in which the state’s largest school district has fought to dismiss a teacher for poor performance, a laborious process that begins with an unsatisfactory evaluation.

Judges sided with the teacher in both cases, meaning the teacher goes back into the classroom unless the district is willing to endure the time and expense of an appeal.

Yet teachers seem just as frustrated with the system. In a survey of nearly 900 Denver teachers, fewer than 40 percent agreed their evaluations were either accurate or helpful.

Of those surveyed, fewer than one percent had been placed on a remediation plan – the typical result of an unsatisfactory rating – in the past five years.

“The evaluation is meant to be used as tool for improving instruction and instead we use it as a final exam,” said Kim Ursetta, president of the Denver Classroom Teachers Association.

“We have principals who wait until now – the final weeks of the school year – to go in and perform observations and fill out evaluations.”

One smaller district south of Colorado Springs is seeing results after increasing the required classroom visits by principals. Superintendent Mike Miles insists his principals observe teachers eight to 16 times per year, far above the one to two visits required by state law.

Denver’s data picture

In 2007-08, DPS principals and assistant principals gave unsatisfactory ratings to 33 out of 2,185 teachers evaluated – or 1.5 percent.

That’s one of the highest percentages of unsatisfactory ratings given recently by any metro district.

In 2006-07, seven out of 1,915 DPS teachers received unsatisfactory evaluations, or .4 percent. In 2005-06, it was 10 out of 1,916, or .5 percent.

State law sets out some evaluation guidelines, such as requiring new teachers to be evaluated every year and tenured teachers every three years. School districts and teachers’ unions usually add to those guidelines in their collective bargaining agreements.

For example, Denver administrators must schedule a formal classroom observation with a teacher in advance and must meet with the teacher beforehand to discuss the lesson to be observed.

But that’s just the beginning. The evaluation flow chart in a district handbook consumes an entire page, complete with arrows linking nine different steps. The remediation flow chart takes another page.

Brad Jupp, a former union leader who became senior academic advisor for DPS, likens the charts to “Fortran programming.” (To see examples, here is the Evaluation Chart and here is the Remediation Chart from the district’s evaluation handbook.)

Jupp, now on loan to Duncan’s office to work on teacher effectiveness, said the complicated schematic evolved because quality teaching can be difficult to measure. Unions wanted to ensure their members were evaluated fairly and their livelihoods protected from biased evaluators.

“In the absence of good information about what teachers did with kids, and in the presence of people having grave consequences committed upon them for arbitrary or capricious reasons, people created a system with a lot of brakes,” Jupp said. “And the brakes have taken on a life of their own.”

Teachers’ concerns are not without merit. In a survey of more than 100 DPS principals and assistant principals, 17 percent admitted they had given a lower rating to a teacher than was warranted. The top reason for doing so was “interpersonal interactions with the teacher.”

The principal and teacher surveys were administered by The New Teacher Project, which is scheduled June 1 to release a national report on teacher evaluations and dismissals. A draft obtained by Ed News shows Denver’s numbers are similar to three other districts studied.

“DPS evaluation processes appear to be approximately as ineffective in accurately assessing and improving teacher performance as those in place in the other three districts,” the draft states.

Few know this better than teachers themselves. In the survey of DPS teachers, 30 percent said there were tenured teachers in their schools who should be dismissed for poor performance.

Of the 50 Denver teachers who received unsatisfactory evaluations in the past three years, most retired, resigned or, in the case of probationary teacher on annual contracts, were not rehired. Seven are back in the classroom after successfully completing remediation plans.

Eight teachers initially fought DPS on their dismissals but most resigned or retired before their cases went before a judge. In the one case that went to a hearing, the judge recommended the teacher be returned to the classroom.

Dismissal battles in Jeffco

In most Colorado school districts, teachers in their first three years on the job work on annual contracts and administrators can simply choose not to re-hire them, citing cause.

That’s how DPS, for example, listed 130 dismissals of new teachers in the past three years. Some new teachers received unsatisfactory evaluations and went through remediation but most did not.

It’s when teachers have tenure, gained after three years, that the lengthy legal fights begin.

State law specifies the grounds for which tenured teachers can be dismissed, such as felony conviction, neglect of duty, insubordination and unsatisfactory performance, the latter category added in 1990.

Colorado Department of Education officials don’t track how many teachers are dismissed in a year or in which category. The data collected by Ed News shows larger districts push for unsatisfactory teachers to retire or resign and only rarely end up before a judge.

Consider Jeffco schools, where Stevenson has filed dismissal charges against four teachers in her seven years as superintendent. The district has 4,700 teachers.

Two dismissal cases involved conduct – a teacher who emailed nude photos to other employees and a teacher who refused to take an alcohol test at school, though a doctor later said she was most likely suffering from acute alcohol withdrawal.

A judge agreed the teacher in the email case should go. But in the alcohol case, in which the teacher admitted to binge drinking “on weekends or during vacations, but never on a school day,” the judge recommended she be retained.

”There is great frustration,” Stevenson said of the cases. “We have tried but at a certain point you become, ‘Why should we go through this?’”

Jeffco also lost the two cases in which it tried to remove veteran teachers for poor performance. The most recent, which began in fall 2007 and ended in April, involved a 10-year elementary teacher and a new principal.

The teacher describes his teaching style as “different” in documents obtained by Ed News under the Colorado Open Records Act:

“He admits that he prefers to allow the students freedom to walk around in the classroom, getting out of their seats more than traditional teachers … He allowed students to choose their seats, sit on the floor, work with their hands, do frequent group objects and guide their own learning, all of which he refers to as progressive teaching.”

What the principal saw was a chaotic classroom and students getting hurt. Parents complained after their children came home with red marks from being pinched by classmates or had their faces written on with highlighters or were “attacked” by other students. Two months into the school year, a student was transferred out of the room at the parents’ request.

Beginning with a Nov. 2 email and continuing through the 2007-08 school year, the principal documented no fewer than 10 class observations, letters or meetings with the teacher. She noted concerns about students wandering the halls without passes, one student choking another while left unsupervised in the classroom and students wandering away from the playground at recess.

The teacher was put on a remediation plan at the end of the school year and continued teaching another semester before being placed on administrative leave in December 2008. The principal, in five classroom visits, noted students yelling at each other, calling each other by ethnic slurs, eating antibacterial wipes and even spitting at the teacher.

The teacher declined an offer from his union rep to resign and went through a three-day hearing in March, before an administrative law judge. The judge, in her findings of fact issued April 15, said the district failed at proving the teacher had made no improvement. In fact, the judge said, the teacher had made some improvement.

“We just got back a decision that I just can’t believe,” Stevenson said after receiving the judge’s ruling. “And the board just acted on it last night and retained the teacher. They didn’t have a choice.”

She acknowledged the district could have done some things better – some exhibits introduced at the hearing contained incorrect dates, for example, and the teacher did not receive an evaluation on the date he was told to expect it.

But the overall result of such decisions is chilling, Stevenson said.

“I will tell you that you are not going to proceed on a performance dismissal with somebody that is just not great,” she said. “It’s not going to happen.”

Fixing a broken system

Efforts are underway in some districts and schools to improve the evaluation system.

Ursetta, the Denver union president, is part of a DPS team of teachers and administrators who have been working since November on a proposal they’ll submit this summer to the Bill and Melinda Gates Foundation for funding.

Team members have visited New York City; Montgomery County, MD; Toledo, OH; and suburban San Diego to look at different ways of mentoring, supporting and evaluating teachers.

“You should be having continual conversations about your instruction, whether your students are benefiting from your instruction, and different techniques and strategies on how to meet content standards,” Ursetta said. “A lot of times, teachers don’t get any feedback from their principals. That’s something we hear continually.”

Rob Stein, the principal of Denver’s Manual High School, agrees a good evaluation system “isn’t just about accountability, it’s about support and growth.”

He and his staff won autonomy from district and union rules and, this summer, they’ll work on creating an evaluation system that includes a clear job description, tasks and responsibilities, and annual goals.

“Principals are still teachers and they’re still caregivers and they’re in relationships with people and they have to work with them and count on them,” he said. “It is very hard to sit down in a meeting with someone who’s a close colleague and tell them honestly, you’re not meeting the standards.”

Clear and measurable goals make that easier, he said, because the process becomes less subjective.

Few, if any, Colorado districts use student achievement data in evaluating teachers. Lawmakers this year created a “teacher identifier” system that, for the first time, will allow the state Department of Education to track individual teachers and their impact on students.

But the Colorado Education Association, the statewide teachers union, asked for and received a provision in the law prohibiting the use of that data to “negatively sanction” individual teachers.

DPS’ Jupp said it’s a question of when, not if, such data will be used in teacher evaluations.

“It’s going to be impossible for us to disregard the significance of using student data in the future,” he said. “We are going to have to entertain how to use it. Therefore, it’s a question of when and how.”

In Harrison, Miles began his initiative to boost teacher quality when he took over three years ago. That includes requiring principals to observe new teachers eight times per semester and to observe veteran teachers four times a semester.

“We’re in the classroom all the time,” said Miles, who leads the effort by making 100 school visits a year, visiting four to five classrooms at each stop.

Struggling teachers get oral and written feedback, professional training and remediation, all the steps required by law, he said. After that, if the teacher is still not satisfactory, the district moves for dismissal.

Harrison, a high-poverty district of 11,000 students, this year has successfully removed four veteran teachers, three of them for unsatisfactory performance.

Yes, it’s time-consuming and yes, it’s expensive. He estimates it costs the district $50,000 per dismissal.

“We’re willing to do it anyway because, at the end of the day, our kids deserve it and they deserve us fighting for them,” Miles said. “If you give an at-risk kid who is behind in his proficiency an unsatisfactory teacher two years in a row, you’ve doomed that child to a poor future or jail or worse.

“Everybody knows it but not too many are willing to do something about it, because of the cost,” he added. “I think they’re looking at the short-term financial cost and not the long-term cost of the many years of damage to our kids.”

Nancy Mitchell can be reached at or 303-478-4573.

The COVID-19 outbreak is changing our daily reality

Chalkbeat is a nonprofit newsroom dedicated to providing the information families and educators need, but this kind of work isn't possible without your help.

Sign up for the newsletter Chalkbeat Colorado

Sign up for our newsletter.