Mitigating heterogeneity in the classroom? – Chatbots as support in nursing training
Keywords: chatbots, education, heterogeneity, German as a second language, quality, human computer interaction
TL;DR: Exploring the possibility of chatbots as a way to improve students' learning experience with a special focus on students with German as a second language.
Abstract: Due to the shortage of qualified nursing staff in Germany (https://www.pflegenot-deutschland.de/ (11.0324)), attempts are being made to counteract this shortage by recruiting specialists from abroad (Böhlich et al., 2023). However, migrants already living in Germany also aspire to careers in nursing and therefore undertake nursing training, which leads to very heterogeneous classes in nursing schools. A significant challenge therefore is effectively addressing these students' heterogeneous learning backgrounds with limited resources. Chatbots have emerged as fundamental elements of innovative teaching approaches to address this issue (Kasneci, 2023). The technological advancement of Large Language Models (LLMs) makes interactions with chatbots increasingly indistinguishable from conversing with real people, contributing to their growing popularity. Chatbots could facilitate such interactions, enhance learners' motivation, personalize learning, and improve learning outcomes (Lazarides & Chevalère, 2021; Zhang & Aslan, 2021). “Language plays a special role in nursing training as communication is a nursing intervention.” (Daase/Fleiner, 2024) Researchers stress “the necessity of orienting nursing training towards the increasing linguistic heterogeneity in the group of trainees – not only due to migration.” (Daase/Fleiner, 2024)
In our study, we investigated whether students in nursing training can benefit from working with a chatbot, with a particular focus on students with German as a second language. The study involved 224 students from vocational nursing schools who created and completed medical educational tasks with and without a chatbot's support (ChatGPT, Model GPT-4). In our presentation, we will present the differences between the produced care plans, focusing on linguistic and specialized competence performed by the trainees in both sets of texts (with and without AI). Experts evaluated all two sets of answers. This allowed us to compare the quality of the answers solved by the chatbot to the solutions by the students. The evaluation focused firstly on the professional quality and secondly on the linguistic quality (morphosyntax, use of technical terms, syntax, and textuality). Statistical analyses highlighted that students' quality regarding both fields of interest was significantly higher when using the chatbot. However, they also revealed that students with advanced language skills benefited more compared to those with lower proficiency levels. The results indicate that using ChatGPT in nursing training could enhance the already existing heterogeneity in the classroom rather than mitigate it. We therefore suggest that – in order to make use of the technological advancements provided by chatbots – training in using AI is urgently required for teachers as well as students.
Submission Number: 27
Loading