GothamSchools — daily independent reporting on NYC public schools

My recent writing on the teacher report card issue has brought me back to my original writing on the topic when I received my first data report. I was a little surprised how my initial reaction echoed my feelings a year and a half later. Even though I’ve publicly advocated for the scores to be released, I still agree with my first assessment:

That said, the whole thing has to be taken with a grain of salt. As much as my job has been overwhelmed by testing, I refuse to judge my performance on test scores alone.

Ultimately my job is to prepare my students for a life of learning and success. Tests that measure students’ academic performance are one way to assess my own performance. But I’d like to believe that there are intangible aspects to my job — for example instilling a love of learning and proper work ethic — that can’t possibly be measured quantitatively. I think any attempt to rate teachers without accounting in some way for these aspects of teaching will be fatally flawed.

Rereading this old post from last year, I was also struck by the conversation that followed in the comments. I think they’re worth reading:

Jonathan said:

But when we open the door to this sort of evaluation, it will become in many cases the only evaluation.

There is nothing easier than picking a number out of a report.

Qualitative observation, qualitative evaluation, would take work.

Skoolboy said:

Ruben,

Would you be willing to share your report? If you’ve only been teaching for a year, it’s very likely that the information is extremely unreliable — i.e., your percentile location in the experience-adjusted distribution might be due to chance.

Aaron Pallas
skoolboy2@gmail.com

NYC HS History Teacher said:

Many teachers say this, and I have to believe it will be true

“When they measure teacher performance on student test scores, every kid will get an A.”

And really, does anyone believe the DOE would have an effective way of figuring out whether teachers are fudging grades or not. After all, high scores make them look better. Just look at the abortion the Regents exam has turned out to be.

I responded:

I don’t remember the exact numbers but I believe my ELA percentile was 58 and my Math was 33. I already knew that math was a weak area for my last year though, and I expect this year’s scores to reflect a big improvement. To NYC HS Teacher, I agree partly, but in the case of the ELA and Math scores, I don’t assign them, so it’s impossible to “give every student an A” to raise my grade. That doesn’t mean however, that teaching to the test becomes the standard practice

Skoolboy responded:

Ruben,

One of the things that worries me about the teacher data reports is that teachers might focus on their percentile ranks — 33 and 58 in this case — without considering just how much uncertainty there is in the data that go into those percentile ranks. The report displays the range of percentiles which might be a teacher’s “true” percentile rank, but we’re drawn to the single number that is the best single estimate.

I think it’s quite likely that estimates such as the 58th and 33rd percentiles represent broad ranges of possible values, to the point that the 58 and 33 are statistically indistinguishable. What appears as better performance in ELA than in math may simply be a matter of chance.

I don’t know how useful these reports might be. Teachers who take their work seriously are going to be striving to improve regardless of what the teacher data report says. But if you find it useful, that’s great.

If any other readers would be willing to share the information in their teacher data reports — and anonymously is fine with me — I’d very much like to see them.

Aaron Pallas
skoolboy2@gmail.com

And I responded:

To be clear, I think the report cards are far from perfect, and yes, more than a little annoying. But I’ve always cared about grades, even if I’m had problems with the teacher or their grading system. Even in these cases, I’m the type who will strive for an A. So if nothing else good comes of this report card, at least maybe I will improve my practice to the point where all my students will achieve beyond the predictions of the system.

To which Jonathan said:

But achieve what? Is your job limited to maximizing their scores on the state ELA and Math exams?If you strive to increase your “grade” you will be working to master test prep. Is that the kind of teaching you want to learn to do? What would you be neglecting?

Unfortunately, I never responded to this final, and essential question. I want to do so now with a resounding refusal to let test prep dominate my practice. I understand the fear that the use of test score data will precipitate test prep centered teaching, and that’s why I think it’s essential we have a open dialogue on valid methods of quantitative and qualitative evaluations for teachers. It’s imperative that this discussion includes teachers.

I believe that good readers, writers and critical thinkers can score well on any test they’re given. These are the skills I have always worked to develop in my students. It’s my hope that by preparing my students for the tests in this way, their scores will rise. To belatedly answer Jonathan’s questions, I do worry that report cards predicated on test scores will drive test prep in the classroom. However with the right balance of qualitative evaluations, this practice would be discouraged. In my own classroom, I think it is possible to strive for high test scores (and by extension a high mark on a teacher data report) without sacrificing what I know are the best practices.