Thread by @bjbalas

First day of S&P is on the books! Today involved telling students how assessments work, so now seems like a good time to share my (new) approach for this term. The TL;DR version: We're not doing points. Points annoy everyone and this ain't Hogwarts. Read on for more. <1/n>

First, a word about #ungrading. I REALLY want to like ungrading. I have colleagues who are strong advocates for it and I've seen success stories here and elsewhere about it. I also had a terrible go at trying it in a much smaller class last year - just didn't work at all. <2/n>

Honestly that experience got me pretty low about teaching - about the worst I've felt about it during my career. BUT, I still wanted to try and break the points-based mentality a lot of our students have. So: a compromise. <3/n>

Here are the main principles for all my assessments: (1) Everything is either graded as "satisfactory" or "not satisfactory." (2) You can revise any unsatisfactory assignment to level up, BUT (3) only if you initially turned it in on time OR asked for more time up front. <4/n>

The logic (such as it is) is that students will learn content by trying until they get it right. Revision opportunities also relieve the pressure of acute problems like illness, child care responsibilities, car trouble, etc. <4/n>

A key part of this is to plan for lots of students needing more time and granting that without much worry. I've found that if you give students room to flex, it very rarely turns into the catastrophe we all fear of all the work coming in later and later. <5/n>

These principles apply to all 3 types of assessments: (1) Multiple choice content questions (3/wk). Meant to be easy, and you answer until you get 'em right. Great pedagogy? Maybe not, but also a way to make sure students at least SEE the important facts. <6/n>

(2) Lab reports (1/wk) - Students describe observations from weekly lab sessions, report what didn't work and come up with question(s) the activities raised for them. Largely graded on completion, but if it's really poor I tell 'em to do better. One chance to revise. <7/n>

(3) Problem sets (1/wk). This is the tough stuff. Quantitative problems about models of visual phenomena. Work in groups if you want and two opportunities to revise within 2 weeks of the original due date (w/feedback). <8/n>

Letter grades are based on how much you get done to a satisfactory level. To get a D, you just have to do the lab reports and the MC questions. Want a C? Tack on 5 satisfactory problem sets. A B? Make it 8. An A? 10 of 'em. <9/n>

I'm curious to see how students calibrate to the grade scale. My guess is that there will def. be students who land their B or C and check out. And you know what? OK. If you only want to go so far, why not just be transparent about it and let you shoot for your own goal? <10/n>

Plenty of students do that all the time anyway and I think it's their call to make. This way, they get to that B or C by doing some fraction of the course correctly. Maybe a weird analogy, but it's like doing physical exercise with better form even if its fewer reps. <11/n>

Am I going to drown in revisions? Maybe a little. But, capping the total number at least limits that somewhat. I also never need to spend time negotiating point values, which takes time and is soul-destroying to boot. <12/n>

So, we'll see how it goes, but this is the plan. I'd love to hear comments or questions about this approach and if anyone wants to hear my rant about #ungrading, I will totally share privately anytime! Wish me luck, and stay tuned for more updates. <end>

@threadreaderapp unroll please