Actionable companion · Feedback & scaffolding
Learning with agency: Integration of artificial intelligence performance prediction and learning analytics to improve student learning in online engineering course
AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.
Evidence and context
Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Higher education. Verify the participant population and task before transferring the findings.
Original authors’ abstract, reproduced under the linked source license:
Abstract As a cutting-edge field of artificial intelligence in education (AIEd) that depends on advanced computing technologies, AI performance prediction model is widely used to identify at-risk students that tend to fail, establish student-centered learning pathways, and optimize instructional design and development. A majority of the existing AI prediction models focus on the development and optimization of the accuracy of AI algorithms rather than applying AI models to provide student with in-time and continuous feedback and improve the students’ learning quality. To fill this gap, this research integrated an AI performance prediction model with learning analytics approaches with a goal to improve student learning effects in a collaborative learning context. Quasi-experimental research was conducted in an online engineering course to examine the differences of students’ collaborative learning effect with and without the support of the integrated approach. Results showed that the integrated approach increased student engagement, improved collaborative learning performances, and strengthen student satisfactions about learning. This research made contributions to proposing an integrated approach of AI models and learning analytics (LA) feedback and providing paradigmatic implications for future development of AI-driven learning analytics. Keywords: Artificial intelligence in education (AIEd), Academic performance prediction, AI prediction models, Collaborative learning, Online higher education Highlights Integrated approach was used to combine AI with learning analytics (LA) feedback Quasi-experiment research was conducted to investigate student learning effects Integrated approach to foster student engagement, performances and satisfactions Paradigmatic implication was proposed for develop AI-driven learning analytics Closed loop was established for both AI model development and educational application.
What kind of evidence is this?
Supporting AI learning science · Comparative / pre-post study reported
Stage: Higher education. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.
Study methods · source passage
The research context was an 8-week, online graduate-level course titled “Smart Marine Structure” offered in the summer semester in 2022 at a top-tier university in China. This course centered on the application of artificial intelligence in ocean engineering with particular focuses on the advanced technologies in mechanical metamaterials, structural health monitoring systems, and artificial intelligence in engineering. There were two classes per week over an eight-week period, and each class lasted for 90 min. The course was a completely online course and the online teaching and learning platform was DingTalk. Participants were 62 graduate students (43 master students and 19 doctoral students), majoring in ocean engineering at Ocean College from the university. The students were assigned into 15 groups (four to five students per group), mixed with master and doctoral students in each group.
Limitations and uncertainty · source passage
The third theme revealed the students' attitudes toward online collaborative writing, including four sub-themes, namely sharing of information and expansion of ideas, convenience to record, share, and regulate writing, no limitations of distance, location, or time, and inefficiency of collaborative writing or communication. Students in both the experimental groups (N = 15) and the control groups (N = 11) mentioned that online collaborative writing made it convenient to share information and discuss and integrate ideas. They mentioned that the collaboration made it possible to record, share, and regulate writing and reconciliate the group's division of tasks (The experimental groups: N = 13; the control groups: N = 10). For example, Wu (the experimental group) reported that, “My teammates were responsible…It was straightforward and easy to share documents and links in the collaboration.” Zhang (the control group) said, “We had a clear division of labor, and it is easy to record and share information by screenshots, video recordings and links.” In addition, 6 out of 33 students in the experimental groups thought there was no limitation of distance, location, or time in online collaborative writing, while no students in the control groups mentioned this advantage of online collaborative work. Moreover, it was worth noting that both students in the experimental and control groups had encountered inefficiencies of collaborative writing and group communication, compared to the face-to-face learning environment (The experimental groups: N = 9; control groups: N = 12). Overall, students in the experimental groups and control group held similar attitudes towards the online collaborative writing activities.
Design the system before the pilot
Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.
- Give feedback on the learner’s reasoning, then fade hints. Record whether the learner can complete the next task without that scaffold.
- Measure engagement separately from achievement. Check whether novelty-driven participation persists and whether it translates to independent performance.
A 90-day district pilot
- Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
- Days 16–30 · Design. Choose one disciplinary task. Require a first attempt, a documented AI consultation, a critique of the output and an independent explanation. Align the comparison group and scoring rubric with the same objective. Name an accountable owner and set the baseline, success threshold and budget before implementation.
- Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
- Days 61–90 · Decide. Blind-score reasoning and accuracy on a new task without AI. Distinguish model examination scores, student perceptions and actual learner gains; only the last directly tests educational benefit. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.
The board decision brief
Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.
Attribution and reuse
Fan Ouyang; Mian Wu; Luyi Zheng; Liyin Zhang; Pengcheng Jiao. Integration of artificial intelligence performance prediction and learning analytics to improve student learning in online engineering course. International Journal of Educational Technology in Higher Education (2023-01-17). https://doi.org/10.1186/s41239-022-00372-4
Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.