Students’ experiences and challenges during the COVID-19 pandemic: a multi-method exploration

Likert-type items or scales, as well as open-ended questions, are frequently used in the collection of student feedback. While the analysis of numeric feedback includes many established statistical techniques, systematic analyses of answers to open-ended questions are quite rare due to the challenges of dealing with textual data. In this paper, we leverage the use of pre-trained large language models (LLMs) to extract topics from students’ textual feedback about their experience during the COVID-19 pandemic. In particular, the open-ended questions focused on mental health and remote learning. There were 696 textual responses from 340 participants. Our analysis using a BERT-based pre-trained model resulted in the identification of 13 topics. To further understand these, we also present results from the Likert-type items related to stress, worry and remote learning in relation to demographic characteristics including age, gender and year of study.