Many folks on my campus are clamoring for AI grading tools. Do you have effective strategies or resources that illustrate the dozens of ways that attempting to automate a flawed assessment strategy with LLMs isn't their best idea? I'd be happy to talk more about it with those interested in LLMs and GenAI.

I have been using alternative grading in higher education for over five years. I have attended sessions and read work on how to implement alternative grading in large classrooms but still struggle to figure it out.  This past academic year I taught a class of 135 and 200 for the first time all while applying alternative grading. My biggest sticking point is still wanting to provide strong qualitative feedback and maintain a feedback loop. While I managed this last year, I know it is not sustainable. Any suggestions on how to still leave feedback and allow for revisions when juggling between a total of 200-300 students a semester? Thank you!

Maria T

Do you have any general advice for reducing the instructor burden required to implement SBG? More narrowly, do you have any advice for how to generate pools of substantively different question versions for use in retakes, or any advice for trying to get students to make the most out of their reassessment opportunities so that they do not require as many retakes?

Features I would like:

(1) Integration between Gradescope and Canvas to pass grades to a learning mastery rubric.

(2) OR be able upload a single PDF of all student work in Canvas that is automatically assigned to each student (Gradescope has this functionality - if Canvas could do it then I would use Canvas Speedgrader instead of Gradescope).

(3) Easier way to do revision of work in Gradescope and Canvas that does not reset the grade or require a link to a separate platform with the revised work.

(4) Better learning mastery gradebook view for instructors in Canvas (individual student progress not just mastered/not mastered, overall metrics for each learning outcome).

(5) Canvas student learning mastery view shows their progress towards a learning outcome, not just mastered/not mastered.

In my previous role I used some forms of alternative grading and loved it. I've used specifications-style grading, exam retakes to foster growth mindsets for students, and even ditched a points-based rubrics for lab reports before.

However, at my new institution we have several instructors teaching different sections of the same courses, and I need to meet certain departmental requirements in teaching my courses for consistency and transferability for students (exams, quizzes, and textbook homeworks must be part of assessment, and must be X%, Y%, and Z% of total gradebook, etc. etc.). This makes it more challenging to use some forms of alternative grading, as I want/need to be aligned with my new department's norms/expectations, but have qualms with aspects of traditional points-based grading and see value in implementing alternative grading into my courses.

I'd love to hear from anyone with similar expectations, or any suggestions for navigating this from someone else.

I am curious about whether or not alternative grading impacts (positively or negatively) students with different types of disabilities? I'll be looking out for any resources on that topic :)

For me, I started bringing in some subjective assessments, which are layered on top of specs sometimes. (It's a hybrid method that feels less than ideal.) Are some of you rewarding deeper thought and higher subjective performance through grades or through other means within a spec grading structure?

Hi everyone! I have been using an alternative grading system for a few years, and the majority of students say they prefer it. However, there are typically a few who prefer to receive a traditional grade no matter how much feedback they receive. I always offer to meet with them and discuss what a grade would be if they want to, but no one does. How are others who are using alternative grading systems supporting students who really want a traditional letter grade on assignments? Thanks in advance for any thoughts/ suggestions!

Is anyone else, like me, using the American Statistical Association's GAISE (Guidelines for Assessment and Instruction in Statistics Education) for designing learning targets for standards-based grading in a non-calculus Intro Statistics course (for freshmen/sophomores)? I'd welcome the chance to talk it through with a colleague.

Here is the GAISE report & supporting resources:

• Report: https://www.amstat.org/docs/default-source/amstat-documents/gaisecollege_full.pdf

• Resources (other reports):

(1.) Is anyone else using MyOpenMath as their learning management system and implementing standards-based grading? I'm preparing to use standards-based grading in online asynchronous courses for Precalculus and (non-calculus based) Intro Statistics, and any advice would be appreciated.

(2.) Has anyone already implemented standards-based grading using a MyOpenMath course shell imported into Canvas? My college uses Canvas as their main platform; I haven't integrated my MyOpenMath course shells to Canvas yet, because I'm still tweaking the course, but I'd love to hear from anyone with experience in this.

Thanks for your time!

Any thoughts on responses I can give that don't require a full explanation of alt grading?

I have a regrade period in two of my courses where students can submit missing or partial credit work to be regraded one time for up to full credit. This has been a benefit for both me and my students and I've sort of reconciled my question about by saying that the goal of my class isn't to teach them those "real world skills," it's to teach them the skills specific to my class, but I'm curious what other ways people are approaching this tension.

One approach is using 0 point rubric items with your rating scale as text (E, M, etc.).

The way I do it is numerical, matching the points to the proficiency ratings in Canvas.

For example, my "Achieved" in Canvas is a 3, "Needs Revision" is 2, "Not Yet" is 1, and "No Attempt" is 0.

In this way you can download the grades from Gradescope, download the blank rubric from Canvas, copy-paste into the rubric, then upload to Canvas to populate the Learning Mastery grades.

I show everything step-by-step in my SOP, under "Standards-Based Grading Resources": https://corson.ku.edu/teaching

Does anyone have any experience with this mixed method approach? How did you get started as you started to switch your classes to more objective-based grading? Thank you for any insight you have to offer.

The poster is now available. Please refer to the link below.

https://docs.google.com/presentation/d/e/2PACX-1vRvP9ERJ2oVE9hssOE4NKrPtIcWzXM85XIbH96KJelNOvpjG2m02brF2Vuee0AaGUXM7-JTYhp7tX69/pub?slide=id.g3e9d2b9548e_0_531.

Feel free to ask questions and leave comments for discussion.

Thank you.

I really liked that we saw some data about the success people had with alternate grading. Especially those that compared them to traditionally graded classes.

One concrete thing I am going to apply is to have a makeup day. However, instead of doing it during finals, I am thinking of doing it on the last day of class. This should help with scheduling conflicts

One thing that I am hoping to explore is the use of testing centers to be more flexible with taking exams and the ability to have smaller assessments