87ConclusionIn some cases, the implementation may be very different, even within the same class; for example, the presentation may be a wish-only task due to class size factors.In the assignment category, reading only references to cite a report was deemed highly frequent and difficult （Q49）. This task may be related to reading tasks in the before-and-after class category and seems crucial for EMI in this department. However, the standard deviation of the frequency for this item was found to be relatively high （SD = 1.5）, indicating that the data were scattered around the mean. The inter-pretation of the frequency of tasks in Q49 should be cautious, given that paper-writing tasks （e.g., Q50 and Q51）, prerequisites for Q49, were found to be low-frequency. This result suggests that the implementation of this task varied widely across classes. Another noteworthy feature of this category is that the standard deviations of each task frequency tend to be relatively high, implying that these assignment tasks vary in implementation from class to class, making it challenging to rank the importance of tasks.This study conducted a questionnaire survey on the frequency and difficulty of tasks that undergradu-ate students experienced in EMI offered by the Department of English Language and Literature, School of Education, Waseda University. Doing so, it identified highly frequent and highly difficult tasks. The results showed that the following tasks were the main high-frequency and high-difficulty tasks in EMI: （1） reading specialized textbooks （e.g., chapter reading, grasping an outline, seeking specific information） or references for a report; （2） listening to the instructor’s lecture or audio/video material about specialized content and answering discussion questions; （3） pair/group discussions on the content of specialized textbooks or given questions; and （4） individual/group presentations of specialized content. Overall, the study’s results indicate that receptive skills （e.g., reading and listening） and productive skills （e.g., discus-sion and presentation） are required for undergraduates to succeed in this department’s EMI classes and specific reading skills are crucial for the statistical traits of these tasks.The significance of this study is that by identifying high-frequency and high-difficulty tasks in the department’s current EMI, we demonstrate the English language skills required in the specific context in a data-driven manner without deviating from actual language use in EMI. Implications for English language skills that students should strengthen will benefit the next generations of EMI students and faculty members involved in preparatory courses for EMI, as it will help them to establish pedagogical task designs. Moreover, our findings would prompt that each institution implementing EMI is required to create academic tasks through a needs analysis such as the one conducted in this study, depending on the individual implementation situation.The first limitation of this study is its small sample size. As noted in the method section, several questionnaire responses were not valid data. The reason may be that the questionnaire was administered online, and we could not provide on-site instructions. We attempted to prevent false responses by including a note at the beginning of the questionnaire and presenting the names of several EAP classes that were not the target of the survey, but this may not have been sufficient. Future research should take measures such as
元のページ ../index.html#93