Enhancing student reflections with natural language processing based scaffolding: A quasi-experimental study in a large lecture course
Bibliographic record
Abstract
Multiple studies have shown that scaffolding plays an important role in regulating and enhancing students' metacognitive monitoring and reflections. However, scaffolding students' reflections in large courses is a major challenge. In the current study, we explored how real-time, technology-enhanced scaffolding affects the quality of students' reflections and academic performance. Two major research questions are: RQ1) Do students in the scaffolding condition construct more specific reflections than those in the non-scaffolding condition? RQ2) How do the scaffolding feature, reflection specificity, and the number of reflections relate to students' academic performance? To address these questions, we conducted a quasi-experimental study with a large sample of undergraduate students (N = 1268) in an introductory psychology course. We designed and used a mobile application called CourseMIRROR that prompts students to reflect on what they found confusing and interesting in the lecture. The app uses Natural Language Processing (NLP) algorithms to evaluate students' reflection quality and specificity using a 4-point scale, with 1 indicating shallow reflection and 4 indicating highly relevant or specific reflection. Course sections were randomly assigned into scaffolded or non-scaffolded conditions. Students in the scaffolded condition were provided an app version with the scaffolding feature, while students in the non-scaffolded condition were provided a different version of the app without scaffolding. Regarding RQ1, we found that students in the scaffolded condition wrote significantly more specific reflections on confusing and interesting concepts. For RQ2, results showed that the number of reflections was a significant predictor of academic performance.
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 | 0.010 | 0.018 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.003 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 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".