Log in  |  Search

Teacher Evaluation: Principals, Principles And Power

On the Schoolbook blog of the New York Times, Philip Weinberg takes issue with my two Edwize posts (Part 1 and Part 2) on New York’s new teacher evaluation law. Weinberg is a principal of a New York City public high school and a supporter of the widely circulated Long Island principals’ letter criticizing the New York teacher evaluation law, and he writes that my posts are a response to that letter. On this point, he is simply wrong: even a cursory reading of the posts makes it clear that I did not discuss the letter, but rather set out to provide a comprehensive explanation of the more important and complex features of the new teacher evaluation framework. But Weinberg’s reading of the issues involving teacher evaluation is nonetheless worth addressing, as it brings much needed clarity to the underlying agenda of the principals’ letter. And since the UFT’s stance toward the Long Island principals’ letter is a frequent matter of speculation, it provides an opportunity to explain where we stand.

In his post, Weinberg writes:

My concern about the agreement is that a large portion of a teacher’s evaluation is to be taken out of the hands of principals. I am disturbed by this, not just because I think this will lead to inaccurate ratings and will pressure teachers in unproductive ways (it will), but also because I believe it speaks to a growing distrust of or disrespect for principals. I am surprised that the teachers’ union would trade a principal’s rating for that of a student’s test score, especially given the recent teacher data report debacle. Are most principals less fair or trustworthy than reductive data? I think not.

Weinberg goes on to cite approvingly Mayor Bloomberg’s forthright defense of the NYC DoE’s position that there should be no meaningful and substantive appeals of negative principal ratings on teacher evaluations: “The principals’ job is to decide who’s good, who’s bad,” Bloomberg had said. “It’s their judgment; that’s their job.” Weinberg then asks “Who could disagree?”

We at the UFT disagree.  Weinberg’s complaint – that “a large portion of a teacher’s evaluation is to be taken out of the hands of principals” – is fundamentally correct. The new teacher evaluation framework checks and balances the principal’s role in the evaluation process. And, pace Weinberg and the Long Island principals, that is a good thing.

Here are the new checks on a principal’s discretion and authority:
1. As it now stands, a teacher’s evaluation is entirely a matter of principal discretion and judgment. Much like Louis XIV, the principal can say: l’évaluation, c’est moi. As a result, a New York teacher’s career can rise or fall depending entirely upon whether his or her principal is an experienced educator and a decent human being. Throughout my career, I have been personally blessed with principals who were both accomplished educators and mensches.  But as a UFT Vice President, cases of teachers who are the victims of principals who abuse their power and who know very little about good educational practices come across my desk on a regular basis. No teacher should have to rely upon the benevolence and wisdom of a single individual given absolute evaluative power.

The new teacher evaluation framework does not substitute test scores for unlimited principal power, as Weinberg incorrectly asserts. Rather, it establishes multiple measures of teacher effectiveness. For the measures of teaching effectiveness (worth 60% of the total evaluation), a local teacher’s union can negotiate peer observations and the use of artifacts of teaching such as lesson plans and unit plans alongside supervisory observations, and for the measures of student achievement (worth 40% of the total evaluation), a local teacher’s union can negotiate performance assessments or portfolios, either alongside the value-added measures using standardized test scores or where value-added measures do not exist, for the entire measure.

Why use multiple measures? First and foremost, it is good educational practice. Just as we receive the fullest and most valid reading of our students’ learning when we use multiple measures in their assessment, the use of multiple measures will provide the most complete and accurate reading of teacher effectiveness. But multiple measures also provide ways to limit and check two particular measures that should not become the entire measure – supervisory observations and value-added measures using standardized test scores.

2. Under the evaluation status quo, a principal can use any standard or conception of good teaching that he or she deems appropriate to evaluate teachers, no matter how idiosyncratic or professionally ungrounded it might be. Every year I make a point of teaching a workshop on lesson planning for high school and middle school teachers, in part because it provides me with many different on-the-ground descriptions of the instructional methods NYC public school principals are demanding of their teachers. Far too much of what I hear would curl the hair of any thoughtful, responsible educator. And the more we see of principals who have been fast tracked through Bloomberg’s Leadership Academy before they even mastered the fundamentals of teaching, the more frequently I hear stories of what could only be described as educational malpractice. What is worse, under the current system arbitrators treat whatever standard and conception of good teaching a principal brings to teaching as “supervisory judgment,” which makes it unchallengeable in the 3020a hearing when a teacher is defending their license and livelihood

The new law requires the adoption of a research-based framework for teaching, such as that provided in the Danielson Framework, as the foundation for the measures of teaching performance. The Danielson Framework represents the best professional thinking in the field of education on the essential components of teaching.[1] When implemented faithfully, it creates a common language to talk about good teaching and common expectations of what an supervisor or peer observer should see when observing a lesson. In addition, the Danielson Framework provides a protocol for doing observations that focuses on acquiring evidence and developing an educational conversation between observer and teacher before reaching conclusions and judgments. That is not simply a check on the use of arbitrary standards by principals; it is best educational practice.

3. Under the current evaluation system, the Bloomberg DoE universally turns down ‘U’ rating appeals, refusing to engage in a substantive review of a principal’s rating on an end-of-year rating. As I noted in the second post I wrote on the new teacher evaluation framework, this was a particularly important issue for the UFT, given the increased weight placed on negative, ineffective ratings in the new evaluation law. Weinberg’s approving citation of Bloomberg on this issue is a clear indication of where he stands on the NYC DoE’s kangaroo court of ‘U’ rating appeals.

The February agreement reached in Albany creates two essential checks on this unfettered principal power: first, in cases where principals are abusing their authority and targeting teachers for advocating for their students or engaging in union activism, this agreement creates an appeal of an ineffective rating to an independent board with the power to overturn that rating (the UFT will have the ability to appeal 13% of all ineffective ratings to this board); second, in cases where the principal has rated a teacher ineffective for his or her teaching performance, this agreement establishes an independent validator who, in the year following that rating, will observe the teacher for a minimum of three lessons and report on whether or not the teacher’s performance in the classroom is truly ineffective.

With these checks in place, teachers will receive ineffective ratings and will be dismissed only when their classroom performance is below professional teaching standards, and even with the proper supports, they can not improve.

The Principals’ Letter
While teachers have common cause with the sentiments of the Long Island principals’ letter on limiting the role of standardized test scores in teacher evaluations, that is by no means the only agenda of the letter. The other agenda becomes apparent when, as its very first recommendation, the letter proposes that a value-added measure of student test scores for the entire school become part of a teacher’s evaluation. Such a recommendation does not limit, but actually expands, the role of value-added measures in a teacher’s evaluation; it is reminiscent of the sort of insanity we have seen in Tennessee, and would make a teacher responsible for the test scores of students he or she did not teach.[2] The reason for such a stance? The new evaluation system is for principals as well as teachers, and the value-added component of a principal’s evaluation is for all of the students in the school. For reasons relating to their own grievances with their evaluations, the principals actually propose to expand value-added measures from student test scores for teachers.

As public school teachers and unionists, we believe in the importance of collaboration with principals in the work of providing our students with a quality education. But we would be foolish to ignore the fact there are supervisors who advocate for and defend a position of power vis-à-vis their teachers, including absolute power over our evaluations. When they take such a stance, we will oppose them.

[1] In significant measure, the Danielson framework represents the refinement and further development of the core propositions of the National Board for Professional Teaching Standards. The National Board is a project conceived by the late UFT and AFT President Al Shanker as a means to professionalize teaching. It has the support of both national teacher unions, and its governing body always has a majority of members who are active classroom teachers.

[2] See Michael Winerip, “In Tennessee, Following the Rules for Evaluations Off a Cliff.” New York Times, November 6, 2011.

To solve that, the state is requiring teachers without test results to be evaluated based on the scores of teachers at their school with test results. So Emily Mitchell, a first-grade teacher at David Youree Elementary, will be evaluated using the school’s fifth-grade writing scores.

“How stupid is that?” said Michelle Pheneger, who teaches ACT math prep at Blackman High and is also being evaluated in part based on writing scores. “My job can be at risk, and I’m not even being evaluated by my own work.”




  • 1 Carol Burris
    · Mar 12, 2012 at 12:56 pm

    Your position surprises me, Mr. Casey. I am surprised that you prefer that teacher evaluations be in the hands of kids who take the tests which comprise 40 of the 64 points that teachers need to get out of ineffective. Both in my writing and in my radio interviews I have made it clear that I prefer a system such as Montgomery County’s and Cincinnati’s, which incorporate peer observations. These systems have been very successful. I will continue to advocate for that alternative.
    You are incorrect in your interpretation of our letter regarding scores. Tennessee’s system has gym teachers choose an area for scores. We propose one score for all, including principals, so we work together as a team. This way, if a teacher has the toughest kids to teach, that assignment won’t hurt their personal score. We also suggest that student achievement be 20%, not 40%. We are publicly asking the legislators that the law be amended so that educators’ scores not be public and that the entire awful system be piloted. Our final concern is that the 64 points needed to leave ineffective be lowered so that all parts of the evaluation can have impact. If you doubt any of this, please go to http://www.newyorkprincipals.org, read the letter and please read Dr Welner’s and my article that was published in The Kappan. It is on that site. Mr. Casey, I am a grandma with but a few years left. I am certainly not afraid of APPR. But young teachers should be concerned. I hope you will join us as we lobby the legislature.

  • 2 Teacher Evaluation: Principals, Principles And Power | Edwize - Angryteach
    · Mar 12, 2012 at 6:16 pm

    […] this link: Teacher Evaluation: Principals, Principles And Power | Edwize Recommend on Facebook Buzz it up share via Reddit Tumblr it Tweet about it Subscribe to the […]

  • 3 Remainders: UFT official pushes back against principals protest | GothamSchools
    · Mar 12, 2012 at 9:56 pm

    […] Leo Casey explains the UFT’s skepticism on principals’ petition against new state evaluations. (Edwize) […]

  • 4 Follow the Money
    · Mar 12, 2012 at 10:48 pm

    This defense seems a bit disingenuous, and doesn’t take into account a provision in the agreement: “Teachers rated ineffective on student performance based on objective assessments must be rated ineffective overall.” Many teachers – not just principals – have serious concerns about what the Union is about to agree to. There is even a petition directed to the Union started by teachers here: http://www.change.org/petitions/stop-the-public-shaming-and-unjust-firing-of-teachers with a lot more information that seems to – as of yet – not have been adequately answered by the state Unions.

  • 5 Leo Casey
    · Mar 13, 2012 at 8:57 am


    If you disagree with Philip Weinberg’s reading of the principals’ letter and of the importance of retaining absolute power over the teacher evaluation process in the hands of the principal, to the extent of supporting Bloomberg on his opposition to meaningful appeals of negative ratings, I think you should state that clearly and forthrightly. That would establish common ground.

    I won’t belabor the point by repeating the analysis of the evaluation scoring I explained at detail here, but I have to say that I find your line of argument concerning “40 of the 64″ points to be both confusing and misleading.

    Lastly, what teachers want out of a teacher evaluation system is that they are evaluated on things they can control. Evaluating teachers on the basis of school-wide value-added measures means that they can receive low grades no matter how hard they work or how well they do. It takes what is already a problematic measure — value added — and rather than limiting it, extends it. And in urban settings, with many struggling schools serving communities with high rates of poverty, it would create an extraordinary disincentive for teachers to teach in such a setting. It takes the bad and makes it worse.

  • 6 Robert Whalen
    · Mar 13, 2012 at 12:31 pm

    Mr. Casey fails to understand that some components of the Danielson rubric unfairly penalize teachers who have the most challenging students. In particular, the Danielson Rubric on student engagement makes no distinction between an AP or honors class and a class of repeaters (who have repeatedly failed a subject and often have poor motivation.) The rubric only scores whether “some” or “most” of the students are engaged.

  • 7 Leo Casey
    · Mar 13, 2012 at 2:43 pm


    I disagree. The Danielson framework does not require that engagement look the same for students with different levels of ability and motivation, but it does ask that students be engaged. There is no learning without student engagement: if we say that we are unable to engage students, we say that we are unable to teach them.

  • 8 Teachers and Principals and NY’s Teacher Evaluations | assailedteacher
    · Mar 13, 2012 at 7:30 pm

    […] interesting exchange between NYC principal Phil Weinberg and UFT Vice President Leo Casey raises some interesting questions about New York State’s new teacher […]

  • 9 mike schirtzer
    · Mar 13, 2012 at 8:49 pm

    Mr Casey, Carol has been a strong supporter of teachers and has been opposed to many of the anti-education measures by local, state, and Federal forces. Lets not burn our bridges and try to work together. This is not the time to take on our allies, how about an impassioned post of this manner against President Obama, Governor Cuomo, Mayor Bloomberg. Once again from all my reading on your stance, it appears that you are in favor of new system that without a doubt will hurt educators all across our state.

  • 10 Leo Casey
    · Mar 14, 2012 at 9:01 am


    It is a bit remarkable to find the accusation of disingenuousness coming from someone who, in the same breath, asserts that there is a provision in the agreement when, in fact, it is clearly NOT in the agreement (it was taken from a press release), and who espouses a petition which absurdly claims that value-added measures will be 100% of a teacher’s evaluation.

  • 11 Will Johnson
    · Mar 14, 2012 at 5:44 pm

    Dear Mr. Casey,

    As a teacher who’s been U-rated unfairly and lived to tell the tale, I’m still confused about why the UFT leadership would think that replacing an imperfect system with a terrible one is a good idea. You argue that “what teachers want out of a teacher evaluation system is that they are evaluated on things they can control,” yet you defend an evaluation system that puts teachers’ careers at the mercy of their students’ test scores. As I said before, I am confused.

    Will Johnson