The Future of Education

The acquisition of academic writing skills is an essential part of higher education [1]. Across the disciplines, it is evident that students have a high demand for introductory and support services for methods of scientific work: According to a nationwide representative study [2] by the German Centre for Higher Education Research and Science Studies (DZWH), university students face difficulties when writing papers. 25.7 % of this sample even report great difficulties (see DZHW, 2018). More than a quarter of all respondents (30.4%) rate the existing introductory courses to the methods of scientific work as not particularly useful for their studies (ibid. p. 13). Due to the increasing number and the growing diversity of students, students tend to receive less one to one support or individual supervision when writing their thesis. In turn, this worsens the situation and leads to a poor scientific education and reducing motivation of the students. In our paper, we would like to address this problem and present a technology-supported solution to improve students’ academic writing. Using an AI-supported text generator, students receive a text proposal generated by the AI on a topic of their choice. Students check the text for stylistic issues, verbose expressions or inconsistent argumentation and improve the text according to the standards of academic writing. While editing the text, the AI assesses improvements instantly making the writing process a joint effort. Lecturers provide feedback on the final version of the text which is also provided to the AI. By improving the AI-based writing assistance, students learn to take the professor’s perspective and can better anticipate, how a good academic text should be structured, and which mistakes should be avoided. In addition to improving the writing skills of students, a further aim of the project is to inspire students for technology-supported learning without losing a critical view on AI. Through the creative approach, students experience technology-supported learning as active participation, in which AI has a stimulating and participatory but not a controlling function. To give students an insight into AI from as many perspectives as possible, the project will also discuss critical aspects of AI-supported learning. The implementation of the project will take place via a prototype in courses on Academic Writing at Graduate Centre THOWL.

Keywords: text generator, artificial intelligence, AI-supported learning, higher education.

References:
[1] Prior, P. & Bilbro, R. (2012). Academic Enculturation: Developing Literate Practices and Disciplinary Identities. In M. Castelló & C. Donahue (Hrsg.), University Writing. Selves and Texts in Academic Societies (pp. 19–31). Bingley: Emerald.
[2] German Centre for Higher Education Research and Science Studies, DZHW (2018). Student Quality Monitor – Studienqualitätsmonitor (SQM). Bundesweiter Vergleich nach Hochschularten. https://www.dzhw.eu/pdf/24/sqm_2018_ randauszaehlung_bund_hs-art.pdf