Abstract
In this study, we explore emotion recognition (ER) processes in educational dialogues on civic and social issues (EDCSI) by assessing the accuracy of ER and the factors that affect ER of external readers. Twenty speech-therapy students with linguistic and emotional training were asked to read a written EDCSI in which they had not participated, to rate the emotional intensity in each speech turn, and to explain which markers affected their rating. A positive correlation was found between the discussants' self-ratings and the mean ratings of the readers, although inter-rater reliability was low. The analysis of the reports of the raters showed that they based their rating on markers at the levels of the local turn, the entire discussion, and the broader context of the author's identity. The findings provide basis for a comprehensive framework to identify emotions in EDCSI and to train teachers to moderate and regulate emotions in EDCSI.
| Original language | English |
|---|---|
| Title of host publication | ISLS Annual Meeting 2023 |
| Subtitle of host publication | Building Knowledge and Sustaining our Community - 17th International Conference of the Learning Sciences, ICLS 2023, Proceedings |
| Editors | Paulo Blikstein, Jan Van Aalst, Rita Kizito, Karen Brennan |
| Publisher | International Society of the Learning Sciences (ISLS) |
| Pages | 242-249 |
| Number of pages | 8 |
| ISBN (Electronic) | 9781737330677 |
| State | Published - 2023 |
| Event | 17th International Conference of the Learning Sciences, ICLS 2023 - Montreal, Canada Duration: 10 Jun 2023 → 15 Jun 2023 |
Publication series
| Name | Computer-Supported Collaborative Learning Conference, CSCL |
|---|---|
| ISSN (Print) | 1573-4552 |
Conference
| Conference | 17th International Conference of the Learning Sciences, ICLS 2023 |
|---|---|
| Country/Territory | Canada |
| City | Montreal |
| Period | 10/06/23 → 15/06/23 |
Bibliographical note
Publisher Copyright:© ISLS.