Despite previous research on the use of the first-person pronoun in academic writing, it has rarely been studied in L2 writing and learner corpus research. In this study, the pronoun I was analyzed and compared between native speaking (NS) and Korean nonnative speaking (NNS) corpora of English argumentative writing samples. To identify differences in its discourse functions, three categories (essay commentator, experience provider and opinion provider) were formulated. The findings show that the normalized frequency of the pronoun was higher in the learner corpus. However, the pronoun occurred less frequently within individual essays but was found in more essays. Unlike the NS corpus, the opinion provider occurs more frequently than the experience provider in the learner corpus. For the opinion provider, Korean students usually selected the verb think. The present study suggests the need to develop students’ awareness of the discursive usage of the pronoun and expand their repertoire of metadiscursive devices.
While the number of English-medium instruction (EMI) courses has increased in many EFL universities, relatively little attention has been paid to students who encounter language difficulties in EMI courses. In order to develop language support programs for Korean undergraduate students in foreign professors’ EMI courses, this study collected and analyzed the syllabi of target courses to identify what English language skills are necessary for these courses. Findings suggest that priority should be given to teaching research-oriented English skills such as academic writing skills, with particular emphasis on how to use primary and secondary sources. Furthermore,students need to acquire speaking skills based on an understanding of readings or lectures so as to participate in discussions and to make presentations. Given the greater number of opportunities to interact with foreign professors, it is necessary as well to teach the appropriate mode of speaking to them. This study also discusses the potential of syllabus analysis as a method to complement needs analysis and to develop language support programs for specific contexts.