I AM DEEPLY CONCERNED ABOUT WHAT’S GOING ON IN NY - AND THEREFORE ELSEWHERE - WITH REGARDS TO TEACHER EVALUATION SYSTEMS AND RTTT. THIS POST IS MEANT TO DEFINE A PATH FORWARD THAT INCLUDES STUDENT ACHIEVEMENT DATA BUT MOVES AWAY FROM OBFUSCATED AND HIGHLY QUESTIONABLE VAM MODELS OF “DATA-DRIVEN TEACHER EVALUATIONS.” IF YOU WANT TO READ ABOUT THE CONTEXT FOR MY CONCERNS, PLEASE CLICK HERE AND FIND MY EARLIER POST.
I HOPE YOU’LL LEAVE SOME FEEDBACK AND CONTINUE THIS CONVERSATION. WE CAN TALK HERE OR ON TWITTER.
As always, I want to start my thinking with questions that need answering instead of solutions I’m hoping will fit. In this case, I’m wondering: What expectation(s) can schools actually and realistically have for teachers when it comes to students’ learning, including standardized testing results?
I know this sounds crazy to ask, and many of you are already thinking that I’m about to lower the bar and let teachers off the hook when students don’t achieve, but that’s not where I’m going. I am, though, very hesitant to back value-added models of evaluation that promote paranoia and professional doubt and suffer from flawed formulas, flawed philosophies, myopic accountability plans, and deep obfuscation. I also believe proactive and reactive action, not accountability, ought to be policy’s goal. Are we - as a culture and a profession - looking to blame people for a problem or prompt people to find a means of moving towards its solution(s)?
So, I need to hone the next phase of my initial question… Can schools ask teachers for:
1) an analysis of what the data from required standardized test(s) tell them about their students?
2) evidence of their ability to formatively and summatively assess their students so that they will create data for things they want to know about their students that aren’t portrayed by the standardized tests?
3) evidence that their curriculum and pedagogy continually reviewed and revised so that they are appropriately designed to move students from their current achievement levels to proficiency and mastery of the desired standards for the course?
Why do I like these questions?
1) They reinforce the idea that teachers’ work must be steeped in students’ learning. There is a non-negotiable connection between data, curriculum, and pedagogy.
2) They allow for teachers’ growth over the course of time instead of unrealistically expecting them to be experts at helping each students in each course as soon as they meet. Not doing this would be akin to telling doctors that they’re no good if they have to “run a test” to see what’s going on.
3) They allow for the application of data from local and standardized testing data without suggesting that one is more important than the other. There is no unilateral approach to improving teaching and learning. It’s why testing-based, standards based, and technology based reforms will all struggle until they include a bigger picture.
4) They recognize that it’s teaching appropriately that moves students forward, not “holding high expectations.” Without great teaching, holding high expectations is nothing but a bunch of soundbites.
5) They force conversations between evaluators and teachers and between teachers and students. Currently, VAM models can be - in fact, I believe almost all are - enforced by a computer in a room somewhere. Students get matched with a teacher. Scores go in. Teacher gets a grade in the mail. This model pushes teachers to get to know their students and their needs as well as administrators to deeply get to know their teachers’ work. It truly couldn’t be done remotely.
6) They provide a ton of room for a lot of supportive feedback loops and collaboration along the way. There are no surprises when you’re not hoping to get all of your answers from a one-time event like a major exam.
7) They support the very real idea that teachers are going to struggle and potentially even fail at times. It’s knowing how to recover and who to ask for help that’ll ultimately make them great at their work. If we’re being honest, we all know that that’s a truth for anyone.
Do I, then, believe that there’s a place for students’ actual scores in a teacher’s evaluation? Until I see some serious research about this working well at an aggregated level, I’m just going to say that even if it’s possible, it doesn’t seem to be worth the infighting and political capital it’ll take. It’s not worth the divisive and desperate culture that’s manifesting in our schools today. I also know that I do fully believe that if we get teachers working towards purposeful and researched classroom practices, the achievement we want will happen.
How can we move to this? It’s simple, I think. Let’s take out the pieces of states’ evaluation systems (in NY, it’s worth 40%) that account for testing scores, and instead create a rubric around the three questions of data-based practice I’ve outlined above. We’ll add it to whatever observation rubric is in use and move forward from there.
Glad to read that you agree that the use of VAM as it stands is flawed-not many administrators will admit to that. I'm also pleased that you feel a more proactive stance should be taken in analyzing data and evaluating teachers based on that data. The blame game needs to stop.
ReplyDeleteYour questions raise a lot of valid points re. how data can be used to more accurately measure student progress and how data can become a more useful tool in evaluating teachers.
I think we need to bear in mind, however, that how teachers collect and interpret data can be very subjective, especially if it's not from standardized testing. In early childhood, we tend to rely on more informal, soft data than formal assessments.
Also, while data is connected to curriculum and pedagogy, curriculum often varies from school to school or district to district. In NY, while Pearson has become a main mandated curriculum, not all districts/schools use Pearson entirely. My school uses Pearson for Math, which is actually not so bad; but we use UBD Units of Study for ELA/Writing which had become an absolute nightmare.
Not only does curriculum vary, but pedagy varies from teacher to teacher depending on experience and education. All teachers teach differently, use different methods/strategies, etc.
Even the use of a rubric to evaluate teachers can be very subjective, especially as to how administrators would use it to evaluate teachers. My administration uses the Danielson rubric to the umpteenth degree regardless of whether they're observing a kindergarten teacher or a fifth grade teacher. There is no room for flexibility or individuality in my school when it comes to teacher observations.
We would need to be evaluated using a fair rubric that pertained more to what grades we teach, as well as respecting the pedagogy of every individual teacher.
I still think it will never be easy to fairly use data to evaluate both students and teachers but I think your argument makes some good points.