Academic writing skills enable students to convey their understanding and critical thinking. However, in many contexts, writing skills are not promoted or training measures are rather ineffective (Thaiss & Zawacki, 2006; Stevenson & Phakiti, 2014). Recent advances in natural language processing (NLP) and machine learning (ML) make it possible to analyze the writing quality of texts (Handschuh & Davis, 2016). This can be leveraged to provide students with individual and adaptive feedback as well as to support gains in students’ writing motivation and writing quality (Rapp & Kauf, 2018; Strobl et al., 2019).Whereas automated support for revisions on the micro-level targeting factual knowledge (e.g., grammar, spelling, word frequencies) is well represented, tools that support the development of writing strategies and encourage self-monitoring to improve macro-level text quality (e.g., argumentative structure, rhetorical moves) are rare (Strobl et al., 2019). Therefore, the goal of our study is to design and evaluate a ML-based learning support system, which is able to assess the quality of generated academic texts and provide students with self-monitoring reports on their writing practice. In this way, we intend to foster students’ academic writing skills, including relevant students’ data literacy as a prerequisite for the effective implementation of such an automated feedback system (Yu & Liu, 2021). In order to investigate long-term effects and change in students´ writing performance and practices, we plan a longitudinal study spanning four years. The expected outcomes of the planned project are the following: A) designed and evaluated data literacy inventory for freshmen students’ academic writing; B) operationalized and validated quality indicators of AI-enabled learning support; C) measured impact on how to improve freshmen students’ data literacy for academic writing and the quality of academic writing, and long-term measured effects on changed writing practice; Overall, the impact provides evidence on how to improve the acceptance of the system, including gained insights on the influencing factors, in particular gender and cultural factors.
Last updated:27.01.2023
