Corpus Construction for Creation of new Writing & Speaking Rating Scales for the Entwicklungsverbund Süd-Ost
- Hosting organisations
- University of Klagenfurt
- Responsible persons
- Nikola Dobric
- Start
- End
With the planned shortening of Austria’s teacher training programme (Bachelor of Education) from four to three years as of 2026, the education of future English teachers faces significant challenges. Despite the reduced duration of study, it remains essential to ensure that prospective teachers maintain a high level of language proficiency (C1–C2). Reliable tools for assessing productive language skills—particularly in writing and speaking—are therefore of central importance. This project addresses this need by developing new, empirically grounded rating scales that respond both to current educational policy changes and to the possibilities offered by modern technologies such as AI.
Information about the Project
The project is based on a multi-step empirical process for developing rating scales:
- Collecting authentic student language productions (written and spoken).
- Expert ranking of these performances according to quality.
- Collaborative analysis and description of typical features.
- Creation of the scales and linking them to the Common European Framework of Reference for Languages (CEFR).
A digital learner corpus will be built, containing at least 300 language samples (150 spoken, 150 written) from students enrolled in English teacher training programmes at the Universities of Klagenfurt and Graz. The data will be produced predominantly by native speakers of German and will consist of academic essays and extended speaking tasks at C1–C2 level.
The corpus work will require technical resources such as a database, parser, manual annotation tools, and an interface for speech-to-text transcription. Server infrastructure will be provided by the University of Klagenfurt, and formal management will be handled by the Klagenfurt Language Testing Centre, which is committed to ensuring the sustainable use and long-term open access to the data. In addition to supporting rating scale development, the corpus will also serve as an important resource for researching the status of L2 English at Austrian universities and for informing educational policy decisions.