For the past seven years, I have served as president of the New Haven Federation of Teachers. In that time, our union has received national attention for partnering with both the superintendent's office and the mayor's office to improve the New Haven Public Schools.
Part of that work has involved creating a teacher evaluation system that treats teachers as professionals and provides those who are struggling with support. While many press accounts have described our contract and the teacher evaluation system it established as "groundbreaking," few have unpacked the details of what makes the system unique. I'd like to share how our union and our members worked with the district to create a teacher evaluation system of which we are all quite proud.
In August 2009, we began negotiating our contract, set to run from September 2010 to June 2014. We had our typical negotiations team of roughly a dozen members. There was a good cross section representing the entire membership: high school, middle school, and elementary school teachers; a guidance counselor; a social worker; and support staff. We also included teachers at all steps of the salary scale. Everyone was represented.
Back in February 2009, the mayor and superintendent met with me to propose a major school reform effort in New Haven. The focus would be on teacher evaluation, turnaround schools, accountability, tenure, work rule changes, and compensation. This effort was clearly something my union members and I were interested in pursuing, as this was an opportunity for us to have real input in improving the school system and to be treated as equal partners in doing so.
To that end, we created a 12-member Citywide Reform Committee: six members were from management, including representatives from the superintendent's office and the mayor's office, and six members were teacher representatives, including four officials from the New Haven Federation of Teachers (Executive Vice President Tom Burns, Executive Secretary Pat DeLucia, Executive Board Vice President for High Schools David Low, and myself). The other two members of our team were Sharon Palmer, who was president of AFT Connecticut at the time, and Joan Devlin, who was a member of the national AFT's educational issues department at the time.
That summer, the regular negotiations team worked on the salary schedule, medical benefits, class-size issues, and other working conditions, while the Citywide Reform Committee worked on the school reform initiative, including teacher evaluation. We negotiated the contract and school reform initiative simultaneously, because the reality (or fear) was that if we did not come to an agreement on the school reform initiative, we would have trouble getting a contract in place by October as required by Connecticut state law.
I was convinced that for our school reform efforts to have a modicum of success, we needed to abruptly change the content and tone of the discussions concerning the problems with public education. The incessant passing of blame from teachers to administrators to state policymakers put us in a position where public opinion was squarely against all of us in public education—in particular, teachers and their unions. The public was screaming for widespread and repressive changes to teacher contracts and tenure.
To gain a foothold in this debate and reverse the tidal wave of criticism, I publicly said that we, as teachers, must be more receptive to changes in our practice. However, I also said that equally important is top-to-bottom accountability, meaning that all those in public education needed to be more receptive to making profound changes for the good of our students.
By early September 2009, we had hammered out the details of the school reform initiative, including a placeholder agreement to work out a new teacher evaluation system over the course of the year. At that time, we created a citywide teacher evaluation committee, made up of teachers and administrators, to create a new teacher evaluation system. We agreed that this evaluation system (known as TEVAL) would be in place by the time the new contract was set to begin in 10 months (September 2010).* We then passed the school reform initiative off to the negotiations team.
To create buy-in for TEVAL, I communicated regularly with all union members. I wrote an article for each edition of our bimonthly union newsletter, updating our members on the process and progress. Then, as developments continued to occur during the year, I shared this information with the teachers via email. We also provided monthly updates at our stewards' meetings and executive board meetings. We spent an awful lot of time making sure we kept teachers informed every step of the way.
A few months ago, we finished negotiations for our new three-year contract, which runs from September 2014 through June 2017. The focus this time shifted from TEVAL to professional development opportunities for teachers, since the heavy lift for creating the new teacher evaluation system was completed in the prior contract. We do, however, continue to make changes and modifications as necessary each year.
Unquestionably, the most significant part of our work in creating TEVAL was replacing the reliance on high-stakes testing to measure student growth and teacher effectiveness with "multiple measures of assessment." It is particularly satisfying that our work here in New Haven has created a ripple effect throughout the nation, as other school districts and states are beginning to use multiple measures of assessment in place of standardized testing. We could never support or accept an evaluation system that relied solely on high-stakes testing. We agreed that standardized tests are useful tools to provide data to drive our instruction. But we remained steadfast in our position that they were designed to tell us what students know and don't know. They were never, ever intended to evaluate a teacher's effectiveness.
Student learning growth, based on multiple measures, such as state and district assessments, teacher-created assessments, and student portfolios, accounts for roughly half of a teacher's evaluation. The other half is based on a combination of instructional practices, such as classroom management, delivery of instruction, and professional values.
For the student learning growth piece of TEVAL, teachers typically write two to four student learning objectives in conjunction with an administrator, referred to as an Instructional Manager (IM). Goals must be mutually agreed upon and data driven. Most often, these goals consist of one mathematics goal and one literacy goal. Having the goals based on student assessment data is essential. However, the student learning objectives are specific to individual classrooms and are not based on district-wide or school-wide data. We set it up this way because no two classes are the same, even within the same school building. For example, in a school with four sixth-grade classes, the data in three of those classes might clearly indicate that improvement in reading comprehension is a priority, while the other class's data might show that reading comprehension is already strong. Therefore, each teacher looks at his or her own students and writes goals based on those students' data.
For the instructional practices piece of the evaluation, which accounts for roughly 40 percent of a teacher's evaluation, we created a rubric for administrators to use when conducting classroom observations. The rubric gives much-needed guidance and provides teachers with clear, objective, and measurable performance indicators. Members of the TEVAL committee spent nearly a year crafting this rubric. They did so with lots of input from teachers throughout the district.
We publicized the names of the six teachers on this TEVAL committee and encouraged our members to contact them with ideas, concerns, and questions. A "working group" was formed to assist in the writing of this teacher evaluation rubric. Approximately 40 teachers joined the working group, which met at the union office twice a month, every month, for an entire year, with additional meetings in between. The teachers on this working group received feedback from colleagues in their respective school buildings, thereby incorporating the ideas of hundreds of teachers. The working group did the actual writing of the rubric and then handed it off to the TEVAL committee to review.
Administrators use the rubric when they conduct full classroom observations. Under TEVAL, they are required to do so at the beginning, middle, and end of each year. Additionally, they typically conduct several 15- to 20-minute classroom walk-throughs so that by the end of the year, the IM has been to the same classroom on numerous occasions. A fair and comprehensive evaluation requires multiple visits, with timely feedback, occurring at regular intervals throughout the school year.
In an effort to reduce the amount of paperwork a new evaluation system can create, we have moved from generating hard-copy paper forms to putting all evaluation data online. Everything is now in an electronic system whereby teachers and their IMs can simply log in to their accounts. Both can view what information each of them has entered, respond where appropriate, and make any updates as the school year progresses. Now meetings between teachers and IMs are much more productive because information has been shared online prior to their formal sit-downs.
Ensuring a Fair Evaluation
In New Haven, we have principals and assistant principals who are outstanding. They are excellent instructional leaders and run their buildings well. However, as in all districts, we also have building administrators who are less than effective. Given that the continuum of school administrators runs from highly effective to downright ineffective, we wanted to ensure that teachers were being evaluated properly and fairly.
In previous years, teachers very often were not made aware that there was a performance problem until April or May, and then they had only until June to show improvement. Now, TEVAL requires the IM to notify a teacher by November 1 if he or she feels that the teacher may potentially be rated as "needs improvement" at the end of the year. This designation must be driven by classroom observations conducted in September and October (and perhaps dating back to the previous year). Once the teacher is notified that he or she may potentially be rated as "needs improvement," a plan of improvement with tangible support is written. Examples of support may include having the teacher watch an instructional coach model a lesson, attend a professional development workshop targeted to his or her particular need, or observe the classroom of an exemplary teacher.
Additionally, we needed to guard against poor evaluations from IMs who may be unskilled in evaluations. We also needed to prevent unsatisfactory evaluations due to personal problems that may have occurred between a teacher and an IM, as well as problems that might arise from individual biases. These concerns have been addressed by a unique system of third-party validators: educators from outside our school district. They are principals, superintendents, and instructional coaches who have resumés indicating outstanding abilities and experiences in the area of teacher evaluation.
Today, the district has approximately 15 validators on contract who have been hired through an interview process conducted by our union and the school district's central office. Both our union and the central office had to mutually agree on all the validators who were hired, and I personally sat in on each and every validator's interview.
Each validator observes three lessons spaced throughout the year with an IM. Both the validator and IM use the same rubric and forms when conducting the observation. The validator does not share his or her report with the IM. At the end of the year, if the IM rates a teacher as "needs improvement," a central office administrator and I review the validator's report to determine if it confirms or refutes the IM's observations.
In the three years the evaluation system has been in place, we have had more than 40 teacher nonrenewals and potential terminations reversed, in large part, by the validator's report. It has proven to be a very powerful tool in protecting our teachers, and it also protects the school district from losing good teachers. It is difficult for urban school districts to attract and retain good teachers, and the validation system serves both purposes.
Also, some of our teachers had their nonrenewals reversed because the district did not provide the necessary supports to help them improve. As I explained earlier, our new evaluation system requires the district to support struggling teachers.
As it turns out, quite a few of those 40-plus teachers were nontenured. In Connecticut, teachers earn tenure after four years of successfully teaching. Elsewhere in our state, a teacher without tenure would have no legal recourse to being nonrenewed for a negative evaluation, as a nonrenewal is at the superintendent's discretion.
At the same time, tenured teachers retain all of their rights and protections as provided by state law. Perhaps the most significant part of our evaluation system is that all teachers, tenured and nontenured, are evaluated under the same system and in the same way. What the validator does is provide an important check on arbitrary decisions made by IMs. This safeguard in the teacher evaluation process has provided my colleagues in the union leadership, as well as myself, much needed peace of mind.
After three years of TEVAL, slightly less than 2 percent of teachers have left the system each year due to performance issues and a lack of sufficient improvement. None of our cases have had to go to arbitration. Throughout this process, both the school district's central office and our union have acted with a great deal of integrity. The district leadership has agreed to reverse the nonrenewals of teachers improperly evaluated for one reason or another. And for our part, we have engaged in the difficult conversations with colleagues who, despite having been fairly evaluated and properly supported, did not improve sufficiently to remain in the classroom.
It is important to remember that the authors of TEVAL did not design the evaluation system for only the teachers at the "needs improvement" end of the rating scale. Our union and the school district's central office were cognizant that all teachers have areas in which they can improve. As a result, "teacher development plans" are created on an as-needed basis. These plans often focus on one or two specific areas, such as preparing data and classroom management. The authors of TEVAL recognize that all teachers, even those most skilled, must be evaluated in the same, thorough manner so that the system can identify and support potential areas where teachers need to improve.
As I look back on the past three and a half years, two things in particular strike me as crucial to the successes we have enjoyed to this point. One is the time we invested in this process and how we included all our partners. It was not, nor could it have ever been, accomplished in a hurried manner. Even so, we were all keenly aware of the urgency of the task before us, and we set timetables that we all adhered to. The second crucial element, of course, is teacher buy-in. While we acknowledge that plenty of hard work remains ahead of us, we feel very good about our collective efforts. We do in fact believe it is "our" system.
David Cicarella has been the president of the New Haven Federation of Teachers since 2007 and has been a member of the AFT Teachers program and policy council since 2010. He has been a teacher in the New Haven Public Schools for 35 years. Before working as an instructional coach in the mathematics department for six years, he taught algebra, reading, and science at Fair Haven Middle School for 22 years and also served as the chair of the reading department.
*For more on TEVAL, see www.nhps.net/node/2328.
Reprinted from American Educator, Spring 2014
One Piece of the Whole
Teacher Evaluation as Part of a Comprehensive System for Teaching and Learning
By Linda Darling-Hammond
- When Evaluation Supports a Collective Perspective
- A Glimpse into High-Quality Evaluation
- Criteria for an Effective Teacher Evaluation System
Using Teacher Feedback to Bolster Evaluation
By Ross Wiener and Kasia Lundy
The Professional Educator
A Fine Balance
By David Cicarella
The Mind Shift in Teacher Evaluation
Where We Stand—and Where We Need to Go
By Angela Minnici