Last semester I experimented with giving students the responsibility of grading themselves. I think it went well enough that I plan to do it again next time, though I plan to refine the process based on what I learned.

Motivation

The idea resulted from a talk by Barry Salzberg, the global CEO at Deloitte, to Columbia Business School’s alumni club. He talked about a challenge that since became part of his legacy at the firm. The challenge was to create standards for all Deloitte’s member firms. Member firms operated independently before joining the Deloitte firm, after which they had to meet standards to continue using the brand.
His challenge was that the firms’ independent histories and cultures made them all agreeing to a uniform set of standards impossible. Germany wouldn’t agree to standards appropriate to France, etc. His solution was to have them create their own standards for themselves. It was a gutsy decision to have other people develop what your board of directors would hold you accountable for, but the delegation of authority worked out.
As he spoke I thought, “It’s easy to listen to someone talk about what they did but harder to see how to apply it yourself.”
Less than a week later, as I was thinking about how to grade students whose leadership projects in class were so incomparably disparate, I saw I faced a similar challenge. I wondered if a similar strategy might solve it.

More inspiration

Years ago, touring Science Leadership Academy in Philadelphia, I was blown away at the leadership skills high school students there showed despite having taken on formal leadership classes. I knew a lot of their skills came from how the school taught, not just what it taught. This experiment seemed consistent with that school’s practices.

Challenges

Let’s say the strategy would solve some problems. It still raises others. The tops ones:

• Students might all give themselves As
• Students might lose motivation to do the homework
• Students might stop coming to class
• The administration might fault me
• Students’ different standards might result in inconsistent grades
• The idea might be unusual enough to distract

What it could help

It could help some things:

• Students would all be satisfied with their grades
• Students would take responsibility for tracking their performance and holding themselves accountable
• Students would determine for themselves what they valued
• Students motivation would move from external to internal
• Students who wanted any grade knew what they’d have to do to earn it

What it forced on me

• Talking to students before doing it. I wanted to gauge how students would react so I spoke to past and current students. Several had had the similar experiences of grading peers for group work and grading themselves for averaging in with the instructor’s grade, though never the full grade.
• Talking to faculty before doing it. I wanted to gauge faculty’s thoughts, especially if anyone had seen the idea result in disaster before. No one considered the idea crazy or doomed.
• Talking to administrators before doing it. Since no student or professor raised red flags, I talked to administrators partly for more perspective and also to make sure I had administrative support (aka cya).
• Clarifying their responsibilities and criteria to grade themselves against (the rubric, roughly). This meant thinking through everything I considered important to hold themselves accountable for before it came up.
• A gut check. I had to think this plan through thoroughly since I was sticking my neck out in several ways.
• Structure. I had to create a template for them to email me their grade and why they felt they deserved it. I had to create a structure for them to hold themselves accountable (which ended up being a forum category). I had to work describing the policy into the course to add more value than the time it took from it.

Overcoming the challenges

I came up with several ideas to address the challenges.

• Regarding absences, I stuck with the policy in the syllabus that unexcused absences still reduced their grade by one-third of a GPA point—that is, from an A to an A-.
• Regarding accountability, I told the class that there was a risk if the class had uniformly high marks that the administration would push back. I told them I would defend them, but that they would need to give me backup that I could demonstrate that they deserved the grades with. For that backup, I made two new requirements
  • An Accountability thread on the online class forum where each student would record their weekly effort
  • Each would fill out and email me a template I created that would show their work at the end of the semester
• I also let the class vote on the policy and found general agreement. I said I was happy to meet anyone who felt uncomfortable expressing a different opinion in front of their peers in office hours, though no one took me up on it.
• I prepared and spent time explaining the policy, also taking questions

Results

I described the policy to my class and asked them their thoughts. No one objected to it. I couldn’t know how many of them decided it meant an easy high mark or that they’d have to work harder.
I saw no change in class attendance and participation, nor in their reflection and posting to the online course forum.
I had to iron out many details and explain the policy several times in class.
The grade distribution seemed about the same as my previous semester’s course, though it was in a different subject. Each student’s grade was consistent with what I felt they earned.
No student argued their grade after.
No administrator has objected.
Students seem to have held themselves accountable. They reported on effort I couldn’t observe, like how much effort they put into homework. I have no way of knowing their honesty, though I wouldn’t any other way either.
My workload in grading was roughly the same, maybe 10% less at the end of the semester, but significantly more when implementing the policy during the semester. I think the implementation work this semester was a one-time cost, though.
My type of work changed. Instead of evaluation and judgment, my task was to review their records of personal accountability and offer advice.

What I would do differently

• I started the policy mid-semester. Next time I will tell them at the beginning, including in the syllabus, describing the goals of the policy and showing them the accountability template at the first session
• I will have them tell me their target grade at the beginning of the semester and what amount of challenge to earn it
• The template I had them fill out at the end focused mainly on their work, less on their reflection and what they learned. I will include more of the latter in a future template.
• I researched as much as I could before committing. I intend to keep researching.

One thing I won’t likely do differently is to adjust for self-assessment biases, which I have been told exist along various axes, for example that high-performing students often assess themselves more critically than lower-performing students. I’ll continue researching this area, but I believe that such biases are comparable to instructor-assessment biases and significantly smaller than the challenges of comparing wildly different leadership and entrepreneurship projects.
I also consider it reasonably that students grading themselves differently reflects their different values. One of the main goals of this policy was to have students challenge themselves on their values—for example how they should weigh work, amount learned, class participation, etc—and move away from me telling them what they should value.

Summary

I experimented applying a successful business strategy of delegating authority and having people hold themselves accountable to an academic context, adjusting as best I could for the differences.
The experiment hit no disasters, nor did I see student effort decrease.
Students seemed to hold themselves accountable.
I’d love to hear your thoughts, especially if you have or know of relevant experience.