Predictive models trained on one semester's offering of a course identified at-risk students in the subsequent semester with high prediction accuracy¶
Claim · All claims
Evidence · 1 study · designr2·q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person · recheck by 2028: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? Random-forest models trained on trace and academic data from a prior semester achieved AUC scores above 0.8 when applied to identify at-risk students in a new cohort. → Wei Dai 2025
Evidence¶
Wei Dai 2025¶
Wei Dai, Jionghao Lin, Flora Ji-Yoon Jin, Yi-Shan Tsai, Namrata Srivastava, Pierre Le Bodic, Dragan Gašević, and Guanliang Chen. (2025). Learning Analytics for Early Identification of At-Risk Students and Feedback Intervention. Journal of Learning Analytics, 12(3), 102–125. https://doi.org/10.18608/jla.2025.8735
q2 · i? · design · r2
Post-hoc evaluation of random-forest models trained on week 0–4 and week 0–7 data from the 2022 source semester and applied without retraining to the 2023 target semester of the same course. The article reports "a high prediction accuracy (with AUC scores above 0.8)" on the new cohort; no effect size is printed.
"our predictive models demonstrated a high prediction accuracy (with AUC scores above 0.8) when applied to a new cohort of students"
Discussion¶
Related Claims¶
- Attendance, behavior, and course performance (the ABC indicators) reliably predict which secondary students are at risk of dropping out — the general claim this finding bears on, with the studies behind it
- Random-forest models trained on a prior semester showed significantly lower AUC when tested on a new semester's data without retraining — reports the opposite
- Early-semester LMS interaction data predicts AI-assisted cheating risk in the final exam, with Logistic Regression achieving AUC = 0.763 under LOOCV — related
- Students not on track for college enrollment and persistence can be classified with about 90 percent accuracy using a small set of predictors — related
- Random forest algorithms yielded the best classification performance in K–8 MMLA studies comparing multiple machine learning models — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
cross-semester-at-risk-prediction-high-accuracy |
title |
Predictive models trained on one semester's offering of a course identified at-risk students in the subsequent semester with high prediction accuracy |
description |
Predictive models trained on one semester's offering of a course identified at-risk students in the subsequent semester with high prediction accuracy |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2025 |
recheck_by |
2028 |
generated |
by process:wiki-ingest on 2026-10-09 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — wei-dai-2025 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.