Academia.eduAcademia.edu

Figure from:Empowering Teacher Learning with AI: Automated Evaluation of Teacher Attention to Student Ideas during Argumentation-focused Discussion

See full PDFdownloadDownload figure
Figure 3: Pyramids for the two multi-transcripts used to evaluate the classifiers that predict whether the given utterance can serve as a justification for the indicator score. Each circle represents an utterance picked as a justification by at least one of the 12 human raters; tier placement is described in the text. The letters denote the first name initials of the five simulated students (Jayla, Will, Emily, Carlos, and Mina) and the teacher (T). Pyramids for indicators b and c are merged for ease of visualization; the sub-pyramid with just students’ utterances is used for indicator b and the sub-pyramid with the teacher’s utterances is used for indicator c.

Figure 3 Pyramids for the two multi-transcripts used to evaluate the classifiers that predict whether the given utterance can serve as a justification for the indicator score. Each circle represents an utterance picked as a justification by at least one of the 12 human raters; tier placement is described in the text. The letters denote the first name initials of the five simulated students (Jayla, Will, Emily, Carlos, and Mina) and the teacher (T). Pyramids for indicators b and c are merged for ease of visualization; the sub-pyramid with just students’ utterances is used for indicator b and the sub-pyramid with the teacher’s utterances is used for indicator c.

Related Figures (4)

in the transcription, each utterance was time-stamped and iden- tified by the speaker (either the preservice teacher or one of the five student avatars). In an earlier study, findings indicated that the preservice teachers had significant growth in their ability to facilitate discussions from the beginning to the end of the semester across five dimensions of this practice [31]. Figure 1: Teacher Interaction with Upper Elementary Stu- dent Avatars in the Simulated Classroom. Image courtesy of Mursion, Inc.
in the transcription, each utterance was time-stamped and iden- tified by the speaker (either the preservice teacher or one of the five student avatars). In an earlier study, findings indicated that the preservice teachers had significant growth in their ability to facilitate discussions from the beginning to the end of the semester across five dimensions of this practice [31]. Figure 1: Teacher Interaction with Upper Elementary Stu- dent Avatars in the Simulated Classroom. Image courtesy of Mursion, Inc.
Table 1: The distributions of the number of utterances per transcript and number of words per utterance per each dis- cussion participant
Table 1: The distributions of the number of utterances per transcript and number of words per utterance per each dis- cussion participant
Figure 2: The distributions of evaluation rubric scores: Indi- cators a, b, c, and total score.
Figure 2: The distributions of evaluation rubric scores: Indi- cators a, b, c, and total score.
Figure 4: RQ 1 experimental design pipeline. The green boxes represent datasets. The white boxes represent machine learning models.
Figure 4: RQ 1 experimental design pipeline. The green boxes represent datasets. The white boxes represent machine learning models.