This is a dataset collected in an Introduction to Data Science course at a university in Australia.
The dataset comprises comprehensive trace data from students using the FLoRA Engine. Unlike prior SRL studies that emphasise brief, tightly controlled tasks, our two-week assignment captures how SRL unfolds across multiple sessions in authentic higher education contexts. This longer window reveals transitions from initial task definition, goal setting, and strategy selection to active monitoring, iterative revision of problem-solving steps, and reflective adaptation—offering new insights into the dynamics of long-term SRL.
In addition, detailed chat logs with GenAI illuminate learners’ information-seeking behaviours, a cognitive–metacognitive process involving goal setting, strategy selection, evaluation, and adaptation. The logs show students articulating intentions and information needs, refining query specificity or scope over iterations, and evaluating responses to adjust subsequent strategies. These interactions provide direct evidence linking information seeking with SRL processes such as planning, monitoring, and adaptive control.
Comparisons between high- and low-performing students further highlight meaningful differences in SRL enactment. Higher performers exhibit clearer goals, more strategic querying, and more effective monitoring and adaptation, while lower performers show more superficial planning, weaker monitoring, and less adaptive behaviour. These contrasts pinpoint targets for instruction and scaffolding to strengthen SRL and information problem-solving skills.
Finally, because participation in FLoRA was voluntary, the dataset includes both users and non-users of GenAI. This contrast enables evaluation of GenAI’s influence on information-seeking strategies and academic performance, clarifying its educational value and informing the design of supports that foster effective self-regulation and improved learning outcomes.
@article{li2025ips_dataset,
title={Dataset of GenAI-assisted Information Problem Solving in Education},
author={Li, Xinyu and Yang, Kaixun and Wei, Jiameng and Cheng, Yixin and Ga\v{s}evi'{c}, Dragan and
Chen, Guanliang},
journal={arXiv preprint arXiv:2601.12718},
year={2026}
}
@article{li2025floraengine,
title={The FLoRA Engine: Using Analytics to Measure and Facilitate Learners' own Regulation Activities},
author={Li, Xinyu and Fan, Yizhou and Li, Tongguang and Raković, Mladen and Singh, Shaveen and van der Graaf,
Joep and Lim, Lyn and Moore, Johanna and Molenaar, Inge and Bannert, Maria and Ga\v{s}evi'{c}, Dragan},
journal={Journal of Learning Analytics},
year={2025}
}
@article{li2025flora,
title={FLoRA: An Advanced AI-powered Engine to Facilitate Hybrid Human-AI Regulated Learning},
author={Li, Xinyu and Li, Tongguang and Yan, Lixiang and Li, Yuheng and Zhao, Linxuan and Raković, Mladen and
Molenaar, Inge and Ga\v{s}evi'{c}, Dragan and Fan, Yizhou},
journal={Computers & Education},
year={2025}
}