I have been reading a PDK article by Paul Black and Dylan Wiliam entitled, "Inside the Black Box: Raising Standards Through Classroom Assessment." A colleague and I have been talking quite a bit about formative assessment and she recommended this article. A few excerpts from the article hit me as I was going about my usual teaching responsibilities. The first one helped solidify all of the thought I've been giving to self-assessment and metacognition on this blog.

"Thus self-assessment by pupils, far from being a luxury, is in fact an essential component of formative assessment...if formative assessment is to be productive, pupils should be trained in self-assessment so that they can understand the main purposes fo their learning adn thereby grasp what they need to do to achieve."
I started to pat myself on the back a bit, but then quickly tasted a slice of humility pie as I read on...
"...the choice of tasks for classroom work and homework is important. Tasks have to be justified in terms of the learning aims they serve, and they can work well only if opportunities for pupils to communicate their evolving understanding are built into the planning. Discussion, observation of activities, and marking of written work can all be used to provide these opportunities, but it is then important to look at or listen carefully to the talk, the writing, and the actions through which pupils develop and display the state of their understanding..." (Emphasis mine)
In my effort to be more "data-driven" in my formative assessments, I feel like I've been a bit too analytical and quantitative. Rather than simply touting standards-based reporting as the end-all solution, I realize I need to listen to the students discuss their misconceptions rather than just seeing the level of understanding as some sort of number on paper. I decided it was worth giving a try.

"Listening" in action
I gave a brief quiz yesterday in one of my math classes. From a standards perspective, it assessed two of this chapter's learning targets. One was a "new" learning target and one was a learning target that was assessed via a quiz once already last week. Today, I handed back the quiz and matched up students according to their relative strengths and weaknesses on their current level of understanding on the learning targets. This was my previous practice. Today I took it a step farther. I circled the room and helped groups who were struggling and listened in with the intent of getting a better grasp of their current misconceptions. This led to a more informed and meaningful whole-class discussion after the groups had time to share with each other.

I believe that listening is allowing me to take a step forward towards answering a key question the article posed:
"Do I really know enough about the understanding of my pupils to be able to help each of them?"
Formative assessment guides instruction
In addition, it led me to reaffirm my belief that an assessment is not "formative" unless it truly guides instruction.

I need to use this quantitative and qualitative "data" (by taking a detailed look at the mistakes students made on the quiz and listening to their conversations as the students talk about their mistakes) to decide how to proceed with instruction even more in the future. Am I transparent in the way I communicate this with students or do they think a "quiz" is merely a "mini-test?" I wonder if we too often separate "instruction" from "assessment" in our classrooms. "Today is a quiz day..." when in fact, the two entities should not be mutually exclusive. I came up with a new motto. It's not really earth shattering and pretty simple:
Everyday is "formative assessment day"
It is more than a change in semantics. Our assessment and instruction should be a cyclical process if we're truly using "formative assessment" techniques. In computer programming lingo, it looks something like this:
START
IF STUDENTS < "GET IT" THEN INSTRUCT
ASSESS
IF ASSESS >= "GET IT" THEN "DONE" ELSE GOTO START
I'm not 100% satisfied with this syntax, probably because there's no one easy way of figuring out the "GET IT" part. Standards-based grading and listening are two possible components, but the puzzle doesn't seem to be solved yet. What other components or key characteristics do you see as essential in effective formative assessment?