On-demand Performance Assessments vs. The Grading Load

One of the things I love about teaching writing with a Workshop approach is the opportunity to support students as they are in the midst of independent projects. This just-in-time teaching allows for differentiation, re-teaching, stretching, and confirming. There are often “ah ha!” moments. Workshop also allows the opportunity for creating a community of learners: peers helping each other, sharing with partners, small groups, or the whole class. Taking a risk in such a supported classroom feels do-able, and students make big growth. Often as I grade final projects, I can see the tracks of my teaching—both whole group and individual—reflected in student work. This is very satisfying as a teacher.

CC0 image from pixabay website

On the other hand, as I grade final projects, I wonder how much of the work is their own. How well can they actually do the things I am asking them to do without all the support they get? Where is the line between scaffolding the learning so students can grow, and expecting them to show their learning independently?

Our new units from the TCRWP staff recommend on-demand writing both before and after a unit is taught. Before-unit writing allows the teacher to see what skills students already have and areas for growth. After-unit writing shows what students have learned from the unit. In-between these on-demand writing pieces is the workshop: self-selected independent projects supported by directed teaching, individual conferring, and peer sharing.

Jay McTighe and Grant Wiggins advocate for performance assessments that show transfer of learning goals. In an Understanding by Design (UbD) white paper (ASCD, 2012), they explain the third of their seven key tenets: “Understanding is revealed when students autonomously make sense of and transfer their learning through authentic performance. Six facets of understanding—the capacity to explain, interpret, apply, shift perspective, empathize, and self-assess—can serve as indicators of understanding”.  The words “autonomously” and “apply” jump out at me. They seem to be agreeing with the TCRWP staff that on-demand performance is the only way to see true understanding and learning.

So what do those ideas mean for me as a teacher? And how does grading intersect with these ideas? Although ideally I’d like to teach and assess without putting a grade on a piece of work, I know that is not the reality of school. Grades are supposed to show how well students are achieving the learning standards. We are obligated to communicate a student’s achievement on report cards, and at our school, that means assigning a grade to their work.

If I take McTighe & Wiggins’ and TCRWP’s suggestions and meld them with my grading obligation, ideally my unit should look like this:

1. Before-unit on-demand writing to pre-assess student learning. This would be used by the teacher to inform upcoming teaching, as well as by students to set learning goals.
2. Three to four weeks of workshop during which students work on self-selected independent projects supported by teaching and feedback.  This would be filled with formative feedback as students practice their new skills within the supported environment.
3. After-unit on-demand writing to assess mastery of learning standards and transfer of understanding. This would be a summative grade that goes on the report card.

I have a few questions about this “ideal” unit assessment scenario. First and foremost, I wonder if students will put their full effort into their weeks-long processed workshop piece knowing that it doesn’t “count” toward their grade. Right now, most students will pour over their writing product with a fine-tooth comb, comparing their work to the rubric descriptors, asking questions of peers and me to fix it up to meet the standard. What’s the point in putting in all that effort if that piece of work doesn’t “count” toward a grade?

Second, what if the post-unit on-demand assessment falls on a bad day for the student? This is one of the big caveats to relying on standardized testing data to define student ability: it’s just a snapshot of one day, and maybe that was a bad day for that student. How can one day’s work define the learning of a whole unit? What if the students know more than the assessment allows them to show?

Third, who has time to score three pieces of writing for every unit? I spend hours evaluating student writing as it is, and I’m not doing all the on-demand pieces. I’ve asked this question in workshops, and get the answer that the on-demands (especially the pre-assessment) shouldn’t take all that long because the point is to just get a “sense” of where the student is, not examine it with a fine-tooth comb. Well, maybe that works in elementary where writing is brief, but give an 8th grader 45 minutes to write an on-demand piece and I get two pages typed—single spaced, size 12 font. Just reading the piece takes at least 5 minutes, and then another 5 minutes or so to score, multiplied times 45 students, and I’m looking at a minimum of 7.5 hours. If I’m assigning a summative grade, as I would for the post-unit on-demand assessment, I would probably double that because I do need to examine the writing with a fine-tooth comb.

Could there be a compromise? What if…

1. Only the student uses the pre-assessment on-demand piece to make goals for the unit? Once the student shares those goals with me, I can use them to guide my conferring, and possibly whole class instruction as well if enough students have the same goals. Once the unit gets rolling, I would focus on the work in front of us and not the on-demand anyway when I am conferring with students. This would take one of the 7.5-grading-hour commitments off my plate.
2. The processed piece and the post-unit on-demand each count 50% of the summative unit grade? This would give enough weight to the processed piece that students would want to work hard on it over the three-four week unit, and also show what students can do independently.  It would honor the long-term work as well as the one-day snapshot. I’m still looking at double the grading time, though. That’s still a problem.

It seems there is a disconnect between the “ideal” and the reality of the day-to-day life of a teacher. I’m not sure how to resolve this. Any thoughts out in the blogosphere? How do you handle on-demand assessments? How do you find time to evaluate multiple pieces of writing for the same set of learning standards?