In Memory of the HC Transfer Grade System
A deep dive into the evolution, complexity, and eventual sunsetting of Minerva's foundational concept transfer grading system.
In light of Elena Cura’s submission, we realized it was time to do a full overhaul of how Minerva’s grading system has changed. ( All the details of the old systems can be found on MyMinerva - but be warned: it is quite a bumpy ride. )
Minerva set out with ambitious visions for transforming higher education. That meant turning every aspect of it upside down. Of course, that also meant that you could not simply finish your course, get your grades, and move on. How could that make the foundational concepts habitual?
In an attempt to foster a growth mindset and incentivize the incorporation of Minerva’s special Habits of Mind and Foundational Concepts (HCs) beyond the first-year courses, the HC transfer grade system was born.
In this system, first-year grades were continuously updated until graduation. Based on key insights from the science of learning, students had to keep applying the HCs throughout their four years.
Now, a decade later, this system is in its final semester. With the M26s graduating, no class at Minerva will have tasted the transfer grade system.
The original system
The first cohorts at Minerva (M19-M23) received their cornerstone grades based on a complex calculation of a mean score, a transfer score, and a transfer scope score (see Box 1). These grades were updated at the end of each semester and not finalized until graduation, meaning that, until then, their transcript showed a “Pass” instead of a letter grade.
Importantly, 3s, 4s and 5s were aggregated together, all counting as “successful” transfers.
However, the transfer scope score was adjusted for the number of courses you had completed, and required continuous applications throughout students’ time at Minerva.
Box 1
HC Grand Mean (60%): The average of your scores across all HCs introduced in the course.
Transfer Competence Mean (15%): The average transfer competence score for all HCs introduced in the course. Computed by finding the net number of successful transfers, adjusting for completed courses, and using the sigmoid function to map it to the 1-5 scale (for more details, search for “transfer” at MyMinerva).
Transfer Scope (25%): This measures the breadth of successful transfer within each cornerstone. It is calculated based on the number of HCs you’ve successfully transferred (that is, getting a 3 or above). You needed about 40% of HCs successfully transferred to get an A.
| Net Successful Transfer Score Benchmarks | 2.75 (B-) | 2.95 (B) | 3.15 (B+) | 3.35 (A-) | 3.55 (A) |
|---|---|---|---|---|---|
| 2nd Year (14 courses) | 5 | 5 | 6 | 6 | 7 |
| 3rd Year (22 courses) | 6 | 7 | 7 | 8 | 8 |
| 4th Year (28 courses) | 8 | 8 | 8 | 9 | 9 |
How many "successful" scores did students need? The table represents weighted scores, so a single score on a 5x project contributed 5 scores to the net count.
Additionally, originally only “non-foregrounded” HC applications counted towards the transfer score. That meant that if the professor tagged (“foregrounded”) an HC with the assignment, it could not count as a successful transfer application, and would rather count towards whatever course it was graded in. The idea was that you should discover opportunities for transfer yourself, and actively engage with a range of HCs.
Although this goal of “far transfer” does occur when applying knowledge in new contexts, note that there is nothing in the literature suggesting it needs to be unprompted.
Transition system
When changing the system in 2024, this was cited as one of its main weaknesses. It came with additional issues too: Rewarding only unprompted HC applications often led to less relevant and more superficial applications.
To avoid these tradeoffs, non-foregrounded HCs were also included in the transfer calculations in the transition system for M24-M26.
Box 2: Example Adapted from MyMinerva
Consider an example. A student has 10 scores (8 @ 3 and 2 @ 4) on #systemdynamics. This results in an HC weighted average of 3.2, which is a B+. Suppose that these scores are all from foregrounded applications, so do not count toward transfer. The student now earns another score: a 3 on a background or unanticipated application. The HC weighted mean score drops to 3.18, but this is in the same letter grade tier (B+) as 3.2, so their grade is not hurt. At the same time, the score counts as a successful transfer effort and raises their Transfer Scope and Transfer Competence scores, pushing their overall composite grade upward, toward an A-. (The exact amount of the change will depend on what other transfer scores the student earned.)
At the same time, the outcome index dashboard was not updated to reflect these changes (presumably, they didn’t bother investing in this as the system was in its final rounds), so students could no longer accurately keep track of their scores during the semester.
This was likely part of the reason the academic committee adopted a so-called do-no-harm policy, where students could not go down from an initially higher grade. For M24-M26, if students achieved an A in a cornerstone (either from first year or through transfers), their transcripts would be updated to reflect this and locked, helping students in grad school and job application processes.
Impact
The Column has not been able to access specific data on how grades changed for M19-M26 from the end of their first year to graduation. In a conversation with the university’s Data Analyst, John Cote, who handles much of this work, he explains that students’ grades usually dipped in the second year, but then picked up and improved overall, especially as capstone and manifest came in with huge weights in year 4.
Cornerstone lead Prof. Terrana explains that while they examine grades for each cohort throughout the year to ensure consistency across class sections, they do not focus on patterns across years. However, consistent with John Cote, she believes that “on average, Cornerstone grades for M19-M26 increase[d] each year (from transfer)”.
Looking ahead
Reading Elena Cura’s (p. 14) article makes it clear that it is quite a vulnerable system. However, these issues could also be the result of bad implementations of an otherwise good idea. Besides traveling, Minerva’s pedagogy is what really sets it apart from its competitors.
Although HCs are now moved into the upper year courses directly, students are no longer rewarded for trying to improve HCs they struggle with, and the end of the transfer system might mean the end of their special status as the key-stones of Minerva’s pedagogical system.
If we don’t just want to be a travel school that chooses tango over textbooks, we need to carefully balance innovation with building systems that work in practice.
Author