M-Powering Teachers - Computationally identifying funneling and

Journal article

Sterling Alic, Dorottya Demszky, Zid Mancenido, Jing Liu, Heather Hill, Dan Jurafsky
arXiv preprint arXiv:2208.04715, arXiv, 2022

DOI: 10.48550/arXiv.2208.04715

Link <<

Cite

APA Click to copy
Alic, S., Demszky, D., Mancenido, Z., Liu, J., Hill, H., & Jurafsky, D. (2022). Computationally identifying funneling and focusing questions in classroom discourse. ArXiv Preprint ArXiv:2208.04715. https://doi.org/10.48550/arXiv.2208.04715

Chicago/Turabian Click to copy
Alic, Sterling, Dorottya Demszky, Zid Mancenido, Jing Liu, Heather Hill, and Dan Jurafsky. “Computationally Identifying Funneling and Focusing Questions in Classroom Discourse.” arXiv preprint arXiv:2208.04715 (2022).

MLA Click to copy
Alic, Sterling, et al. “Computationally Identifying Funneling and Focusing Questions in Classroom Discourse.” ArXiv Preprint ArXiv:2208.04715, arXiv, 2022, doi:10.48550/arXiv.2208.04715.

BibTeX Click to copy

@article{alic2022a,
  title = {Computationally identifying funneling and focusing questions in classroom discourse},
  year = {2022},
  journal = {arXiv preprint arXiv:2208.04715},
  publisher = {arXiv},
  doi = {10.48550/arXiv.2208.04715},
  author = {Alic, Sterling and Demszky, Dorottya and Mancenido, Zid and Liu, Jing and Hill, Heather and Jurafsky, Dan},
  howpublished = {}
}

Abstract

Responsive teaching is a highly effective strategy that promotes student learning. In math classrooms, teachers might "funnel" students towards a normative answer or "focus" students to reflect on their own thinking, deepening their understanding of math concepts. When teachers focus, they treat students' contributions as resources for collective sensemaking, and thereby significantly improve students' achievement and confidence in mathematics. We propose the task of computationally detecting funneling and focusing questions in classroom discourse. We do so by creating and releasing an annotated dataset of 2,348 teacher utterances labeled for funneling and focusing questions, or neither. We introduce supervised and unsupervised approaches to differentiating these questions. Our best model, a supervised RoBERTa model fine-tuned on our dataset, has a strong linear correlation of .76 with human expert labels and with positive educational outcomes, including math instruction quality and student achievement, showing the model's potential for use in automated teacher feedback tools. Our unsupervised measures show significant but weaker correlations with human labels and outcomes, and they highlight interesting linguistic patterns of funneling and focusing questions. The high performance of the supervised measure indicates its promise for supporting teachers in their instruction.