Using Möbius for automated assessment in mathematics: a case study
Bibliographic record
Abstract
Abstract We describe a failed pilot that involved using the automated grading software Möbius in place of graduate student markers for three undergraduate courses delivered in the School of Mathematics and Physics in Queen’s University Belfast. We analyze the effects of this change on student engagement and performance. Our evidence suggests that students are more likely to engage with formative assessment activities when they are marked with Möbius. Students also perform better in summative assessments when they have had Möbius assignments to complete—with one module having a stark reduction in failure rate from 32% to 5%. When we surveyed the students who had the opportunity to engage with Möbius, we did not find that they had much enthusiasm for the software. However, we found that students also lacked enthusiasm for the systems for assessment and feedback that Möbius had replaced. Their responses to our survey instead indicating that students may not fully understand the distinction between formative and summative assessment. As we discuss in the conclusion, this project failed because, in spite of this apparent success, we could not drum up the support for Möbius from students and colleagues that justified the expense associated with purchasing software licenses each year. To introduce automated grading in our context we need a system that has zero or negligible associated cost as it will likely only ever be used by a small number of staff.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.020 | 0.075 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.004 | 0.005 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.004 | 0.004 |
| Research integrity | 0.004 | 0.004 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".