“Hey Chatbot, What Do You Think of My English?”: A Qualitative and Quantitative Analysis of Written English Interactions by Italian Students in the UNITE Project

The present research is situated within the framework of the UNITE – Universally Inclusive Technologies to Practice English project, which investigates the use of chatbots to help students practice English as a Foreign Language (EFL), with special attention to the accessibility and inclusivity of these technologies. In broader terms, the present study is situated within the framework of interactional studies and aims to address the lack of corpora of digital communication featuring interactional annotation. Specifically, the present study investigates the extent to which interactional phenomena characteristic of human interaction appear in written human–chatbot interactions. This research also examines recurring patterns to develop a more refined categorization of these phenomena. For this purpose, a sample of texts was annotated, and representative examples were extracted. This made it possible to create a more detailed taxonomy of interactional phenomena, fundamental to a better understanding of how chatbots interact in language-learning-driven conversations. Alongside this qualitative line of inquiry, a quantitative analysis was carried out to investigate the extent to which errors made by students and corrections produced by chatbots are correlated. Therefore, this study represents both a qualitative and quantitative contribution to the UNITE project.