Work

Is it in their words? Teachers' enthusiasm and their natural language in class–A sentiment analysis approach

Year: 2025

Type: article

Abstract: INTRODUCTION: Teacher enthusiasm is an undisputedly important characteristic of teachers, with demonstrated positive effects on student outcomes. Existing research typically operationalised teacher enthusiasm via trait-based teacher- or student ratings. Strikingly little is known about how teachers' trait enthusiasm manifests in their actual in-situ classroom behaviour. Some findings have been reported regarding teachers' nonverbal behaviours, but the links between teacher enthusiasm and teacher language are unknown so far. METHODS: The present contribution fills this research gap by applying lexicon-based sentiment analysis to quantify teachers' emotional tone from transcribed teacher utterances obtained from video recordings of full mathematics lessons (45 min). N = 19 secondary school mathematics teachers and their N = 393 students participated in our study. We realised the sentiment analysis using Remus et al.'s emotion lexicon SentiWS (v2.0, 2019). We obtained both teacher self-reports and student ratings to assess teachers' enthusiasm, shown habitually (trait), and during the videotaped lesson (in situ). RESULTS: Regarding trait enthusiasm, teachers' own, but not the students', ratings were positively linked with teachers' verbally expressed sentiment in the videotaped lesson, specifically its positive valence. Regarding in-situ enthusiasm, associations were even larger but also did not reach significance for the student ratings. CONCLUSION: This is the first study to employ sentiment analysis on transcripts of German teachers' in-class talk. Besides the quantitative links between teacher enthusiasm and their language, it also provides qualitative insights on positive emotional teacher talk in mathematics.

Source: British Journal of Educational Psychology

DOI: 10.1111/bjep.12734

Authors: Anne C. Frenzel, Hannah Kleen, Anton K. G. Marx, David F. Sachs, Franziska Baier‐Mosch, Mareike Kunter

Institutions: Ludwig-Maximilians-Universität München, DIPF | Leibniz Institute for Research and Information in Education, University of Siegen, Goethe University Frankfurt

Language: English

Topic: Sentiment Analysis and Opinion Mining

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Quality education

Open Access status: hybrid

Estimated APC paid: $3,760