Thursday, December 5, 2013

How to do SBG, better, next time!

Wordle of this semester's SBG Targets
Over the last few semesters, I have begun implementing a standards-based grading approach in my classes. My first real attempt was over the summer in a small intermediate algebra class. The results were very positive, but I had fewer than 12 students in the class for most of the semester. Plus it was summer. Could SBG stand up to the challenges of an academic year? Thinking the benefits were worth it, I decided to go for it.

It was worth it -- and I'll talk about why another time -- but I also learned a few things along the way. As the semester is drawing to a close, and I'm grading a lot of student work as they seek to fill in "holes" in the SBG gradebook, I thought it was prime time to reflect on what I would do differently next time: hence, this post.

First, I didn't do a good enough job encouraging students to submit their evidence throughout the semester. In part, that's because I didn't have all of my SBG learning targets prepared in advance -- and that's because another goal of mine was to make sure that each day built off the last in a coherent manner, so I never quite knew where we would be one week out (within reason, of course). I think not asking students to turn in evidence on, say, a weekly basis ended up resulting in a fairly stressful last week of classes for my students, it certainly resulted in a lot of end-of-semester grading for me. I'll fix that next time.

Subset of my SBG gradebook. Columns = targets. Rows = students.
(data was randomized for this example)

But let me say this: the grading I did at the end of the semester -- where students were resubmitting old assignments, and some new ones, in order to show what they had learned -- this was some of the most rewarding grading I've ever done.

Yes, I did just use the word rewarding to describe a grading experience. I actually looked forward to sitting down for a bit of grading. My students were explaining what they had learned and how their evidence showed it, and it was very satisfying to read.

Here are some other forward-looking thoughts about my SBG experience this semester.

Looking ahead to future SBG iterations:

  • If at all possible, I need to have the full set of targets prepared ahead of time. This is not easy your first-time using SBG in a course, but it will be important for the steps that follow.
  • I will do more to involves students in the process of evaluating and compiling their evidence. It will be important to do this early and often. 
  • I will collect smaller tasks more frequently and continue to collect larger tasks occasionally.
  • I've learned you don't need to grade everything on these SBG assignments. The important piece is the narrative and the evidence it refers to. Extra work may support learning, but I will consider what piece(s) represent the evidence for the learning targets. Students can help with this by preparing narratives -- see below.
  • As I assign work as part of the regular class routine (but perhaps not quizzes or other formal assessments), I think I will invite students to turn it in after it has been vetted and revised. There is little point in grading an unfinished product in an SBG system, provided the opportunity to offer feedback is not overlooked. Checking for correctness (vetting responses) can be done--and much learning can occur--through large-group or peer-evaluation in class.
  • I will continue to invite students to mark up their evidence with a narrative that relates the evidence to the relevant SBG targets (that's targets, plural). Students should be able to recognize the places in their own work where they think they are addressing the learning targets. Sticky notes work well for this purpose.
  • I still need to figure out a way to have students record their own evidence so they understand what they have and what they still need to find. Google doc? I don't know.
  • I need to create a more formal structure of "how much evidence is enough". Example: Any target with only one piece of evidence automatically is reduced one level (say, from "4" to "3"). My colleague John Golden used that rule this semester, and I think I like it. Else there is too much ambiguity at the end about "how much evidence is enough?" "It depends" is a very unsatisfying answer, but without a formal system in place early, that's the best you can do.
I'll be asking my students tomorrow to share their own suggestions for SBG, so I'll add to this list once I hear back from them. Stay tuned. And please share your own suggestions in the comments. I'm learning as I go, and I'd love to learn from your experiences too.

Update: I have some new thoughts as I grade the portfolio of evidence students have submitted to me so far. I will add to this list as the grading continues, and then once again when I unseal my students' feedback after final grades are submitted. Stay tuned...
  • I regularly ask students to reflect on their learning, often emailing me those responses. I think I might use that to help students connect with the learning targets in the future by offering reflection prompts like: with regard to learning targets X, Y, and Z, a) What have you learned? b) Which elements of your written work contain evidence of those targets? and c) How convincing is the evidence? [bullet added on 12/8/13] 
  • I need to figure out a more flexible gradebook system, probably one where students record their scores. I'm thinking I score their evidence, which they document in a shared google doc with targets, descriptor of the evidence, and the date submitted. They keep track of it, then submit their top two or three pieces of evidence. I think I would still keep track of evidence from mandatory assignments (quizzes, tests, etc.) in my own gradebook. It's either I do something like that or I need to be a lot less flexible in which assignments I allow as evidence, but that would sort of defeat the purpose. [bullet added on 12/8/13]

    No comments:

    Post a Comment