The Emperor has no Clothes

The NYC DOE recently released effectiveness ratings (TDRs for Teacher Data Reports) for all public school elementary and middle school teachers in NYC. Newspapers and websites immediately published them. The teacher with the lowest rating was identified and publicly ridiculed. The teachers with the highest ratings were lauded.

Some important public figures also commented:

Bill Gates came out against this in an opinion piece for the New York Times: Shame Is Not the Solution. He calls this move to release the data a big mistake, and notes that the feedback did not offer substantive information that would allow teachers to improve.

Diane Ravitch predicted, in the New York Review of Books, a future in which No Student Left Untested. That is, the logical conclusion is for more grades and subjects to be tested, as more data is needed to humiliate more teachers. She also fears that the public humiliation will drive even the best teachers from the profession.

The Math Babe (Cathy O’Neil) noted that Teaching Scores Released, and accurately calls this the “teaching to the test model”. She notes that open-sourcing this value-added model would make it somewhat more useful for teachers, and finally, echoing Ravitch, calls on Bloomberg to study whether the use of this model and publishing the results has “the intended effect of keeping good teachers and getting rid of bad ones”.

Stuyvesant math teacher Gary Rubinstein did an interesting study, Analyzing Released NYC Value-Added Data, and presents some seeming extreme cases of teachers whose rating varies widely from one year to the next, and even in the same year. Very telling are his scatter plots, that make it clear that these extreme cases are not unusual at all.

For those unfamiliar with the abuse of the Value-Added Model, please read the essay Mathematical Intimidation: Driven by the Data, written by Math for America president John Ewing. You’ll learn basic background and limitations of the Value-Added Model, and specific reasons why its application to teaching is misguided and a case of intellectual bullying. Ewing calls on mathematicians to “confront people who misuse their subject to intimidate others into accepting conclusions simply because they are based on some mathematics”.

Closer to the field, I had the pleasure of reading a letter sent home to parents from Liz Phillips, the principal of PS 321 in Brooklyn. In this letter, principal Phillips comes out strongly against the TDRs and in full support of her teachers. She states that these TDRs will actually lead to a larger “achievement gap”, as only standout, high-performing schools like hers will be able to withstand the pressure to teach to the test. You can read a similar message from Ms. Phillips in the blogpost A principal at a high performing school explains why she is “absolutely sick” about the public release of the TDRs.

Given all this, I believe that The Emperor has no Clothes. Showing off the results of this model strips away any pretense that this is useful, or even adds value to the public discussion on the teaching profession. Perhaps this moment is a hidden opportunity to embrace a different way to measure effective teaching that is based on classroom observation and coaching, and not on test scores or a value-added model.

Experienced educators know that good teaching comes in many forms, and they can recognize it in the classroom. I imagine that the effort of developing and articulating a research model of effective teaching (which probably already exists), and organizing an extensive network of experienced educators who visit classrooms to observe and offer guidance to help teachers improve their practice will cost much less than of all this testing. And it just might make a difference.