In recent months, both school districts and teachers unions have agreed that our current system of teacher evaluation is broken -- but just how to fix it has been the topic of bitter debate. In Los Angeles, UTLA, the union representing public school teachers, filed a lawsuit challenging the implementation of a pilot evaluation system that would use standardized test scores to measure student growth. But this month, the union decided to suspend their legal challenge -- at least for now -- and work with the district to improve and implement the new system. This is an important step forward for LA teachers and for LA children. We need an evaluation system that helps teachers like me understand what we're doing right, and what we need to improve upon. Student growth data should be a component of this system.
I moved to Los Angeles eight years ago to become a middle school math teacher because I believed I could make an impact on the lives of young people. In the years to come, my most memorable days were those in which I was able to see that impact taking place, whether it was helping a student who was far below grade level finally master his multiplication tables, or challenging an advanced learner to read and calculate in binary code. Ultimately this is the calling of every public school teacher: to attempt to make an impact on each of the incredibly diverse students who walk through our door.
Until very recently, any sense I had of my own efficacy as a teacher was either anecdotal or pseudo-statistical. Many former students drop by my classroom and express how much they learned there; nonetheless, state test results often showed that the highest scorers leaving my class were the same highest scorers who entered my class. It was difficult to measure exactly what impact my own teaching strategies were having.
In the last few years, school systems have begun to adopt models that are designed to measure how much value teachers add to their students' learning. At present, LAUSD is in the process of piloting a new teacher evaluation system that includes this controversial "value-added" component. The district plans to quantify teacher contributions to student learning through the use of student outcome data, using a model they are calling Academic Growth over Time (AGT). AGT, like most value-added models, is designed to measure a student's growth from year to year in comparison with the growth of similar students across the district. LAUSD hopes to eventually use this measure, combined with additional measures of teacher effectiveness, in teachers' evaluations. The concept of a metric designed to isolate the actual effects of teaching practices is intriguing, not only for me, but for the thousands of other teachers who have always known we've made an impact, but yearn to know how much.
Unfortunately, a lot of the outcry over this new evaluation system misses the point. While no evaluation system will ever be perfect, this should not keep us from moving forward to develop one that actually serves the teaching profession. As teachers, we know the importance of giving our students meaningful feedback on their work -- not as judgment, but as opportunity for improvement. This same attitude should be driving the development of teacher evaluation systems. By having better data available, LAUSD will be able to learn from their own teachers what practices are working and, in so doing, develop more meaningful professional development opportunities.
The conversation regarding how we use student growth data is certainly in its infancy, but it is absolutely the right conversation to be having. And let's be clear: there are still many unanswered questions regarding both the reliability and consequences of using such a measure. While I fully believe in the value of using student growth data, LAUSD must address both concerns around the accuracy of the metric and the potential unintended consequences that a test-based metric can create, such as the incentive to "teach to the test," or even (as we have seen unfortunately in some cases) to cheat. But these issues ought to be starting points for powerful conversations between districts, teachers, families, and the union about not only how we measure excellence, but what we do with the knowledge when we find out.
For example, given these concerns, might the district and the union be willing to use AGT initially as a very small (perhaps 5%) component of a teacher's overall evaluation? In future years, this weight could be increased as the district and union work together to improve the reliability of AGT, to develop systems that ensure the teaching of a deep and rich curriculum, and most importantly, to connect these scores to meaningful opportunities for professional development geared toward helping teachers like me become even more effective.
Making an impact for kids is the reason we joined this profession. A system that provides options and pathways to do exactly that is one that should excite any teacher.
By Kyle Hunsberger, Los Angeles Unified School District math teacher and Teach Plus Teaching Policy Fellow.
Follow Teach Plus on Twitter: www.twitter.com/teachplus
I understand the positions each of you have taken come from a very emotional and personal space. I applaud that. For without passion, our profession would be nothing more than robots dumping information into a child's head (not happening). While I admire and share your passion, I encourage you to take a logical gaze into the current state of evaluation.
While reading the Widget Effect, I wasn't surprised to find that most of the teacher evaluation in our nation states that teachers are doing a wonderful job. My first year teaching was what my students refer to as a “hot mess”. I still received an above average evaluation in my first year. How was this possible? I knew I sucked. My colleagues knew I sucked - granted, they saw potential, but they knew it. Was my boss the only person who didn’t see it? I think my evaluation fell into the culture of “niceness” in schools. The same “niceness“ we afford our kids when we want to help them get rid of that math phobia has crept its way in our evaluation systems. The problem with the “niceness” invading our domain is that there is no room for it there. We are professionals. We are putting our skills and practices into action to help kids learn. They don't have time for our niceness.
Aren’t students worth finding out who is a good teacher and finding ways to duplicate those practices? I think so, and I hope you agree.
Bill Gates should be ashamed of himself considering how many thousands of times my Microsoft Windows programs have crashed, failed, had "fatal errors" or needed constant updating over the last sixteen years. If any of these computers had been a car, they would have immediately qualified for replacement under the California Lemon Law. This approach to teacher evaluation is just another lemon.
There are of course, extenuating circumstances where even this model could be unfair. If a class has an unusually broad spectrum of ability from student to student, a teacher can't be expected to teach all of them effectively. The fault would lie with the administration for forming the classes improperly. If many individual students have problems that begin during a single year and all have the same teacher, that teacher will be unfairly penalized. If a teacher, however, gets low value added metric scores for most of his classes over several years, he is probably doing a poor job.
What I would like to see, if teachers don't like this idea, is how THEY would like to be evaluated.
Like I said below, the teacher submits a lesson plan that explains how it will meet the required criteria (content objective also known as the standard, language objective, how learning will be assessed, details of the plan for the entire period, how the plan will meet agreed upon annual learning goals and accomodations for english language learners and special ed students) at the pre-evaluation meeting where the evaluator reviews the information and reccommends suggested changes.
The evaluator (principal or vice principal) watches the teacher for the period and scores the teacher bsed on the criteria set in the pre-evaluation.
Afterwards, the evaluator and the teacher sit down and review what the evaluator saw and assigns a rating. Later the teacher submits answers to questions in writing and the stack of papers are signed and placed into the teachers file.
The only change I would like to see happen would be that a senior teacher (preferrably in the subject area) perform at least one of the evaluations.
I personally think that incompetent principals and other administrators are a bigger problem than incompetent teachers, so I don't think trusting them to do evaluations is the right strategy. In fact, I would subject principals to the same metrics as the teachers, averaging an entire school rather than individual classes, and of course accounting for demographics.
Do you have any criticisms of the value added metrics themselves, or do you just want to avoid additional evaluations in general?
At least that is the evaluation process I had as a geologist working for engineering firm for ten years. Those that tell you otherwise are either blatantly lieing or haven't worked a day in their lives.
Now that we're moving in this direction (standardeized tests as a measure), they want something else... the bottom line being, they don't want to be evaluated at all! Well that's too bad: You've made your bed, now lie in it! I have NO sympathy for the teachers in this case.
**********************************
What are you talking about?
1. We can't "refuse" to allow evaluations. We have bosses. They have a job to do.
2. Teachers have always been evaluated. I've always had administrators in my room, watching my lessons, asking for my plans, checking into my paperwork, in the 80s, 90s, 00s, and up to this year. So did my parents in the 40s and 50s and 60s and 70s.
3. Anything else you have to say is as probably just as invented. Thanks for the fiction.
Moreover, when you say: "LAUSD must address both concerns around the accuracy of the metric," you demonstrate a true gift of understatement. Can you or your astroturf group produce a single peer reviewed study from a legitimate source demonstrating VAM/AGT being accurate even 85% of the time? Even 85% would be woefully unacceptable for high stakes decisions, but I ask rhetorically knowing that you cannot produce such a study. I write alongside and correspond with most of the experts in this field. I am aware of all the research on VAM/AGT, and the only "work" I've ever seen touting its value are policy papers from reactionary right-wing think-tanks, many of whom share the same funders as your vile astroturf organization, Teach Plus. Coincidence? I think not.
I will wait for your response....
Look, no one's saying that kids don't come from bad home environments. But teachers DO have the ability to affect students in their classroom... you do realize that don't you? This wouldn't be an issue, if teachers simply let their performance be evaluated with subjective, common sense ways. Because they refuse (e.g. because "the administrator has it in for me"), we are where we are at--a bad method of evaluation (yes, I think tests are a small, rigid way of evaluation).
But we have been painted in this corner, because teachers say no to every other method.
Even in-school issues cannot be solved. Tell me how you will evaluate student growth in art class. Explain how you can compare a SPED inclusion class with 12 students, 2 of whom are on IEP's, with a SPED inclusion class of 25 students, with 10 on IEP's. Explain how you will evaluate the "AGT" for the foods class, or for PE.
The only content where you can approximate this kind of evaluation is math, and even there the variables are far too tangled to come up with a teacher number that is accurate even to one digit.
"Value added" requires teachers to accept an unproven evaluation on faith. I'm not willing to put my faith in such a bogus construct.
Teachers will always try to hide from accountability......so it goes.
Neither is the case.
Non-teachers will always try to pretend they know what they don't.
My class was not assigned at random, these students were lumped together so that the special needs teacher and the ELL teacher could provide services to them easier. My scores will be lower because of my student population than other teachers in the same school. Our schools scores will be lower than a school in a suburban area because of the population. Evaluate me on my teaching, not on their scores.
1. My students did an outstanding job on the CST last year. However,I feel like I cheated them out of a wonderful learning experience. Instead, I had to focus more on the state standards than exploring creative writing, poetry workshops, and literature circles. You see, this type of learning doesn't cover the standards. For example, the standards want students to simply identify different forms of poetry. Forget them learning how to create their own poems. No time for that!
2. My students lives are difficult. They come into my classroom with huge issues. I don't care what people like Bill Gates say. These life experiences do affect how they learn.
3. A good teacher KNOWS when their students are learning.
4. There is a lot of research out there that has debunked the effectiveness of creating a testing culture. Please do some research on the Finnish educational system.
5. The Value Added approach is riddled with flaws. And don't even try to tell me that it isn't!
I am really curious to know who has been evaluating you. My evaluations, since effective administrators have been hired, have been full of insightful and constructive feedback.
Teaching them to deal with stress is the reality of most of life.
Quit lying like a rug.
National Partners
The Bill & Melinda Gates Foundation
Carnegie Corporation of New York
Anonymous
Boston Partners
The Boston Foundation
W.C. & J.V. Stone Foundation
Boston Public Schools
Massachusetts Department of Elementary and Secondary Education
Anonymous
Indianapolis Partners
The Mind Trust
The Joyce Foundation
Central Indiana Community Foundation
IU Health
Ayres Foundation
Chicago Partners
The Joyce Foundation
W.C. & J.V. Stone Foundation
Chicago Public Education Fund
Illinois State Board of Education
Memphis Partners
The Bill & Melinda Gates Foundation
Los Angeles Partners
The Bill & Melinda Gates Foundation
The Wasserman Foundation
Washington, D.C. Partners
The Bill & Melinda Gates Foundation
CityBridge Foundation
T3 Partners
EdVestors
Richard and Susan Smith Family Foundation
Massachusetts Department of Elementary and Secondary Education
Shippy Foundation
Anonymous
But it remains reality. Student standardized test scores are affected by other factors much more strongly than teaching, and even the huge classes seen in inner-city schools are too small a sample size to judge teachers on the small influence they've got on the scores. Really, we could accomplish the same result much more cheaply if we just introduced a random number into teacher evaluations.
But we'd be better off evaluating them on whether they're good teachers. Student test scores don't help with that.