license: mit
datasets:
- stanford-nlpxed/classroom_management_data
pipeline_tag: text-classification
library_name: transformers
Model Card for Behavior Management Model
Model Details
- Developed by: Mei Tan, EduNLP Lab @ Stanford University Graduate School of Education
- Release Date: 2025-11-17
- Paper: Tan, Mei, and Dorottya Demszky. (2025). Do As I Say: What Teachers’ Language Reveals About Classroom Management Practices. (EdWorkingPaper: 23-844). Retrieved from Annenberg Institute at Brown University: https://doi.org/10.26300/9yj6-jn52
Model Description
This model is a RoBERTa-base classifier fine-tuned to predict binary labels from teacher utterances in classroom transcripts. It was trained on 8057 annotated teacher utterances from elementary math classroom transcripts from the NCTE dataset [1]. It is intended for research on teachers' classroom discourse.
The model classifies whether a teacher utterance is an instance of behavior management language. Behavior management language is a subset of classroom management language focused on responding to individual or group behaviors that initiate a competing vector ofaction—behaviors that threaten or distract from the instructional trajectory.
Examples include: “I need voices off, and I need your eyes on the board”; “I’m only gonna call on people who are sitting down”; “Everyone needs to be listening”; “Student E, don’t pack up”; “Stop that right now”.
Intended Uses
Not intended for evaluation of teaching quality. What is appropriate in a given classroom is highly contextual and relational in a way that this model does not capture.
Data Formatting
The expected input is a formatted sequence of utterances comprised of the target utterance to be classified and three utterances before and after. The seven utterances should be marked with the respective speakers, and the central target utterance should be marked with [TARGET][/TARGET].
Example: "Teacher: Uh, no, I’ll help you. Student: Somebody wrote w-w-w that [inaudible]. Teacher: All right, everybody get a worksheet. [TARGET]Teacher: I will give you time to make yours, also.[/TARGET] Student: Ooh, yes. Student: Ooh. Teacher: Don’t use the same numbers from yesterday – it’s a different graph."
Generalizability
The training data for this model come from ~200 observations sampled from the original NCTE study [2], which represents 1652 includes observations of 317 fourth- and fifth-grade mathematics classrooms across 53 schools in New England that were primarily serving low-income students of color. The utterances in this dataset are roughly sentence-length and human-transcribed.
Applying this model to new datasets generalizing to other contexts should involve validation: annotate a sample from the new data context to assess model generalizability.
[1] Demszky, D., & Hill, H. (2023). The NCTE Transcripts: A Dataset of Elementary Math Classroom Transcripts. In 18th Workshop on Innovative Use of NLP for Building Educational Applications.
[2] Kane, Thomas, Hill, Heather, and Staiger, Douglas. National Center for Teacher Effectiveness Main Study. Inter-university Consortium for Political and Social Research [distributor], 2022-06-16. https://doi.org/10.3886/ICPSR36095.v4