Here is a copy of the letter that I authored and sent to Chancellor Merryl H. Tisch regarding teacher-performance reviews in New York State. Several other legislators signed on to the letter.
February 3, 2015
Chancellor Merryl H. Tisch
89 Washington Avenue
Albany, New York 12234
Dear Chancellor Tisch:
We believe that New York State’s Annual Professional Performance Review (“APPR”) process fails to accomplish the purposes for which it was developed, and provides unreliable data rather than accountability and transparency. While we believe that teachers and principals should be evaluated by trained administrators and held accountable for their performance, for the reasons set forth below we do not believe that the APPR, as currently constructed, is a reliable measure of teacher or principal performance. Nor do we believe that the APPR process contributes to the professional development of teachers or principals. And although Education Law Section 3012-c purports to expedite the process for termination of ineffective teachers and principals, it actually has the opposite effect.
- The APPR “HEDI” scale is seriously flawed and makes it impossible for an evaluator to differentiate meaningfully among educators. A teacher receives scores on three sub components of the APPR: (i) student growth on state assessments (worth 20 points), (ii) locally selected measures of student growth (worth 20 points), and (iii) locally developed teacher practice measures, mostly classroom observations (worth 60 points). The composite score that a teacher must receive on the three sub components of the APPR in order to be deemed Effective is 75 out of a possible 100 points.
The scale does not work because the percentage of available points required by the state for an Effective score on the first two subcomponents (45%) is much lower than the percentage of available points required for an Effective composite score (75%). Imagine a teacher who receives a 9 out of 20 points on the first two sub components, which deems her Effective. In order to achieve a composite score of Effective, she must receive 95% of the available points on the third sub component, which is negotiated by the school district.
With a very narrow point range to work with, i.e., 57 to 60 points, an administrator cannot meaningfully differentiate among effective teachers through the scoring of the third sub component, the locally developed teacher practice measure.
We urge you to recommend that the Commissioner fix these inconsistencies in the scoring ranges as part of the annual review process.
- The APPR sets up for failure good teachers, good principals, and good schools. A teacher who earns a rating of Effective or Highly Effective may in a subsequent year earn an Ineffective rating because his new class of students did so well in the prior year that there is no room to demonstrate the amount of student growth required for a higher score. Even if his students do extremely well on standardized tests compared to similar students across the state, he may receive a poor APPR score.
A principal too can be penalized for setting high standards for his students. A principal may receive a score of “Developing” if the school does not offer more than five Regents as is required for a higher score. Many high performing school districts give their own, more rigorous exams in lieu of a Regents exam, and thus do not meet the threshold for a higher score. Another criterion for receiving a score of “Effective” or “Highly Effective” is the percent of students scoring 80% or more on the Algebra I Regents in ninth grade. In many districts throughout the state, the stronger students take the Algebra I Regents in eighth grade, and the weaker students take it in ninth grade. The scoring does not account for the students who took and passed the exam as eighth graders, thereby penalizing the principal.
- The APPR is designed for less than 20% of all teachers. Only teachers of fourth through eighth grade ELA and math receive student growth ratings based solely on state standardized test scores. Teachers of other grades and content areas (more than 80%) must work with their principals to agree on student learning objectives (“SLOs”). If a teacher teaches a course that has a state-mandated assessment, such as a Regents exam, that teacher’s SLOs must include as evidence the results on those assessments. Teachers who teach other grades or non-regents subjects will have SLOs that may include test results from third-party vendors, district or BOCES developed assessments, or locally developed assessments—and the target outcomes are determined locally. Although SLOs are treated as if they are comparable in reliability to standardized test scores, they can’t be comparable simply because they are locally developed, with locally determined target outcomes. Even if one accepts—which we do not—the premise that results on ELA and math tests are a reliable indicator of student achievement or educator effectiveness, it is clear that it is unfair and unreliable to compare teachers whose ratings are based solely on state standardized test scores with teachers whose ratings are based on SLOs.
- Most teachers do not receive appropriate professional development based on their APPR scores. A better designed APPR would require principals to work with all teachers, including those rated Effective or Highly Effective, on their professional development. It would require principals, as educational leaders in their buildings, to develop the most appropriate evaluation tools and the best professional development program for all teachers in their schools. New York is missing an opportunity to incentivize our best teachers to become leaders among their peers.
- Education Law Section 3012-c makes the dismissal of ineffective teachers more difficult than under prior law. The APPR is supposed to enable administrators to identify those teachers who are ineffective, and to use the APPR ratings in an expedited 3020(a) proceeding. However, the use of the APPR in disciplinary proceedings requires two consecutive annual ratings of Ineffective, the development and implementation of a teacher improvement plan, and validation of those Ineffective ratings through at least three observations by an independent validator. The result is that ineffective tenured teachers will teach for at least two years before they can be removed from the classroom based on their APPR scores. Even worse, the law makes the termination of ineffective probationary teachers and principals more difficult than was the case under prior law.
New York State’s APPR process fails to accomplish the purposes for which it was developed because it provides a one-size-fits-all approach that does not adequately take into account differences in educator experience, class composition, subject and grade level taught or other factors. It is unreasonable to assume that the same standardized evaluation tool will fairly and reliably rate an experienced PE teacher in a rural middle school, a brand new fourth grade teacher in an urban school, a mid-career guidance counselor in a high-performing suburban school, and a principal of a BOCES career and technical education program. Because the tool is unreliable, the data it produces is also unreliable. Therefore, the proposed increased reliance on standardized testing would unfairly penalize, or fail to identify weaknesses in, teachers, principals, schools and school districts.
We believe that the Regents and the State Education Department must convene a group of superintendents, principals, teachers, school board members, and parents who can advise the legislature, based on their broad knowledge and expertise, how to improve Education Law Section 3012-c so that it will achieve the important goals of increasing educator accountability, encouraging professional development to develop great teachers and principals, and expediting the termination of ineffective teachers and principals.
Thank you for your consideration and for all that you do for the students of New York State.
Members of Assembly
Cc: Vice Chancellor, Anthony Bottar
Dr. James Dawson
Harry Phillips, 3rd
James R. Tallon, Jr.
Charles R. Bendit
Dr. Betty Rosa
Dr. Lester W. Young, Jr.
Dr. Christine D. Cea
Wade S. Norwood
Dr. Kathleen M. Cashin
Dr. James E. Cottrell
- Andrew Brown