Teacher alchemy? The potential for combining digital and social methodologies in supporting learners in the humanities and social sciences

Abstract

Debates surrounding the use of data science in educational AI are frequently rather entrenched, revolving around commercial models and talk of teacher replacement. This article explores the potential for digital textual analysis within humanities and social science education, advocating for a sociologically-driven approach that complements, rather than replaces, the professional skills of teachers. It outlines existing methods for analysing learner writing using socio-cultural theories, including Halliday’s Systemic Functional Linguistics, and Maton’s Legitimation Code Theory. While the quality of linguistic output only forms one learner progress indicator, it is central to the process of schooling, and therefore deserves scrutiny. Digital Textual Analysis methods can be developed through natural language processing techniques to harness both formal and informal learner texts to map conceptual development and language use, informing classroom practice and assessment. The article emphasises the ethical considerations of such data use, advocating for socially inclusive analysis techniques that ensure fairness, transparency, and inclusivity. By leveraging machine learning and statistical techniques, we suggest that digital textual analysis can significantly enhance teachers’ ability to monitor and support learner progress. We present a vision for redefining teacher professionalism in the digital age, proposing a balanced integration of technology that enhances rather than undermines the teacher’s role, ensuring a learner-centred, ethical approach.