This paper examines how large language models (LLMs) can be used to identify, model, and classify topics from responses to the open-ended question included in TALIS 2024. The paper also provides insights into teachers' perspectives by exploring their recommendations to policymakers for improving education systems beyond what closed-response formats, such as multiple-choice questions, can capture. The findings illustrate both the potential of this methodology for educational research and teachers’ recommendations across participating systems.
Forthcoming
Identifying teachers’ recommendations in TALIS 2024 open‑ended responses
Working paper
Will be released on
Share
Facebook
Twitter
LinkedIn
Abstract
In the same series
-
26 August 202647 Pages
-
30 July 202640 Pages
-
Working paper
Exploring practices and evidence for training, supporting and retaining second‑career teachers
16 July 202665 Pages -
Working paper
Concepts, indicators, and practices across OECD countries
19 May 202669 Pages -
3 March 202661 Pages
Related publications
-
9 November 202647 Pages
-
29 October 202637 Pages
-
26 October 202643 Pages -
13 October 202615 Pages
-
26 August 202647 Pages
-
30 July 202640 Pages