教育評論第39巻第１号

20/160

Conclusion14早稲田教育評論　第 39 巻第１号ground knowledge. This becomes particularly time-consuming when the reading assignment is extensive. The time allocated to these courses is always in trade-off with the limited time students have for other class preparations. Support such as handouts with summaries of background information or reading guides could help students focus on understanding the textbook content. For reaction papers, students would appreciate specific aspects to focus on, and for term papers, recommendations on key literature would be helpful. In group discussions, students sometimes had to rely on using L1 Japanese （and even ChatGPT）, simply because it was faster to complete the tasks that way, when there were many questions to answer in a limited amount of time.Our study’s findings provided a better understanding of the difficulties students face within the specific context of an undergraduate semi-structured EMI program at a leading private university in Japan. Throughout the interviews, we also observed a strong wish among students to see EMI classes become even better, enriching their learning experience. Some students expressed that such improvements would benefit them for study abroad or graduate school. Our findings offer insights for more effective pedagogical practices and student support that are detailed and context-specific, addressing the call for more context-specific analysis in EMI research （e.g., Aizawa, 2024; Macaro, 2022; Sahan, 2021）. They may also offer insights into other EMI contexts similar to those described in the Context section of this paper. One insight gained is that the kinds of tasks vary significantly depending on the subject area of the course. For example, in cultural history classes, tasks may involve diverse audiovisual materials such as docu-mentaries and require background knowledge of historical figures. Group discussions in literature classes can also be highly abstract, which can present significant challenges for students. Therefore, EMI courses should thus not be viewed uniformly; instead, repeating Sawaki （2017）, investigations similar to this paper should be conducted for each academic field. Future research in EMI should explore various disciplines to further understand these context-specific challenges and needs. As explained in the Methods section, this study faced challenges in recruiting participants from cultural studies or English literature classes, accessing only one student. The main difficulty was that most English literature classes are conducted in Japanese, which limited the pool of participants relevant to this study. It would be ideal in future studies to gather a larger number of interviewees over several years with the aim of reaching a point where the data collection seems saturated. This would allow us to better identify and understand any unique difficulties that are specific to the subject areas.Methodologically, the approach we took in our research series appears to be effective for gaining a finer understanding of students’ difficulties. Our previous study, Kudo et al. （2024）, involved interviewing students about program-specific tasks and itemizing questions, which were then used to create a question-naire on task frequency and difficulty. Based on these results, in this study, we conducted interviews to explore the nature of these difficulties and the support required. The questionnaire format was instrumental in organizing the interview sequence. However, a limitation of this study is that our understandings of