[CHI 2025] PeerEdu: Bootstrapping Online Learning Behaviors via Asynchronous Area of Interest Sharing from Peer Gaze

The instructions below are old figures. For updated ones in our final manuscript, we will update soon.

Please note that student_demo.csv directly records student demographics from survey. As a result, there may be duplicated student_id if students submit the demographic data for more than one time. However, in other files, there may be new student_id's behavioral data whose student_id is not available in student_demo.csv because some students particiated the study without submitting their demographic data. As such, we suggest you to find overlapped student_ids between student_demo.csv and other data files for simulation, and then make all student_ids unique using set().

To draw figures in the paper

• Use figure_plot.py
• Run f2() in figure_plot.py to draw figure 2 in the paper
• Run f3() in figure_plot.py to draw figure 3 in the paper
• Run f4() in figure_plot.py to draw figure 4 in the paper

To replicate results reported in the paper

• Use result_analysis.py
• Run s1_gaze_manipulate() in result_analysis.py to replicate results in the gaze manipulation subsection in Results section in the paper.
• Run s2_learn_experience() in result_analysis.py to replicate results in the learning experience subsection in Results section in the paper.
• Run s3_learn_outcome() in result_analysis.py to replicate results in the learning outcome subsection in Results section in the paper.
• Run s4_decode_learn() in result_analysis.py to replicate results in the decoding learning process subsection in Results section in the paper.

Cite us

Please cite our paper if you find our datasets/codes useful.

@article{xu2023peer,
  title={Peer attention enhances student learning},
  author={Xu, Songlin and Hu, Dongyin and Wang, Ru and Zhang, Xinyu},
  journal={arXiv preprint arXiv:2312.02358},
  year={2023}
}