You Are Watching:
Dr. Marzano Outlines the Sources for Teacher Evaluation Data
One of the more powerful features of the iObservation system is that a teacher can have feedback from multiple sources relative to their performance in a particular strategy or area if you will. That’s important to have multiple sources for feedback.
Let me give you an illustration. Let’s say I as the teacher have decided that I want to work on my ability to chunk content into small digestible bits over a period of time, and I want to get a feel for how I am doing. One source of feedback to that is my own self rating, but I want more pieces of information. Another piece of information could be from a formal observation that was made of me, and I may cue the observer to give me a rating on the scale as to how well you think I am doing on this area.
Another source could be from a peer if the school is open enough to have peers interact with other peers relative to instructional strategies. You could invite a peer in and say “please look at me relative to this.” Our walkthroughs come into this. If by chance a walkthrough had occurred and they give me a score on this particular area called “chunking” that’s another piece of data. A final area is that you can survey students and a few of the places we have worked at have done that. They’ve actually created surveys asking the students about the teacher’s performance on specific strategies.
With all of those sources of information, I have multiple perspectives on my own performance. A technique I like, instead of relying on any one as the absolute score, you use all sources of information to come up with the best estimate of a teacher’s true score at any point in time.
Let me explain that concept of coming up with a true score. Within measurement theory the score that you receive from any rater, they call it the observer score, actually is only an estimate of what you really deserve. It’s called the true score. The true score is something that’s unobservable. It’s very hard to actually pinpoint, and the more surfaces of data we have the better we can estimate that true score. Another way of looking at this is the teacher has their own self rating, a rating from a peer that they invited in, a rating from a walkthrough, a rating from a formal observation and they surveyed some of their students as to their use of a particular strategy. Now they take all of that information and build a logical case for a certain scorer. Usually what will happen is that they will say “Out of these 5 scores, there are three of these that basically say I one thing, and the other two are maybe different scores.” Well, that’s a logical case right there. There are three of these that say I am at a “two level”. I’m not making the egregious errors, but I still haven’t mastered the strategy yet at a fluent level.
So, a teacher uses multiple data points to then justify their true status at that point in time, and that can be a negotiated score actually meet with a supervisor and say “that’s why I think I’m a two or a three or whatever it is.”