Collaborating with Mature English Language Learners to Combine Peer and Automated Feedback: a User-Centered Approach to Designing Writing Support
Bibliographic record
Abstract
300,000 immigrants move to Canada each year in search of better economic opportunities, and many have limited English language skills. Improving written literacy of newcomers can enhance education, employment, or social integration opportunities. However, frequent, timely, and personalized feedback is not always possible for immigrants. Online writing support tools can scaffold writing development by providing this feedback, but existing systems provide inadequate support when instructors are inaccessible. In this paper, we show how feedback system design can leverage peer and automated feedback to support mature English Language Learners’ (ELL) needs and practices. We identify strong associations between epistemic beliefs and learning strategies, highlighting the importance of tasks that activate productive epistemic beliefs. We find learners accurately assessed high-level issues in a peer’s writing and are accepting of automated feedback, demonstrating that a platform combining peer-review and machine feedback could promote meaningful discussions. We present the results of our mixed-methods investigation that integrates three sources of information: analysis of learners’ psychometric constructs, writing samples to identify error patterns, and participatory design group sessions incorporating human-centred design methods. We synthesize our results into four guidelines derived from seven findings resulting from the investigation of a system that scaffolds writing development for mature immigrant ELLs in the absence of formal instructional support. First, we find that ELLs require a platform to collaboratively iterate through the writing process. Next, we suggest how peer feedback can be enhanced through automated support. We then demonstrate how rubric design can guide both linear and holistic peer-review. Finally, we illustrate why open learner models and learning dashboards should contextualize real world progress.
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 | 0.006 | 0.029 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
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".