Lecturer language: EMI students’ experiences on first-and second-cycle degrees

Previous research has long focussed on EMI lecturers’ English usage as an aspect which might affect lecture comprehension. Fewer studies have directly questioned EMI students about how their lecturers’ language competence affects their experiences in the classroom, and while many studies have focused on postgraduate students, less has been said about undergraduates. We aim to provide further insight into students’ views of EMI lecturer discourse by comparing the experiences of undergraduate and postgraduate students, both local and international, at an Italian university. Data for this study were collected from an online semi-structured survey of 128 students. Findings indicated that it is students’ past experience of English-taught courses as well as their familiarity with non-native English accented speech that influences students’ opinion of EMI lecturers’ language performance and their assessment of intelligibility in the classroom. The findings will serve to highlight difficulties and critical points for further development and pedagogical application.


Introduction
Since the Bologna process, international courses at non-anglophone universities have increased exponentially. This situation, mirrored the world over, has been accompanied by an increase in research into all aspects of English-Medium Instruction (EMI). Previous research has long focussed on EMI lecturers' English usage as an aspect which might affect lecture comprehension but fewer studies have directly invited EMI students to address the issue of how far their lecturers' language competence affects their experiences in the classroom, and still fewer have compared the experiences of students of first-and secondcycle degree courses in an EMI context.
Past research has suggested that greater experience with EMI might affect lecture comprehension. For example, Clark (2017) investigated the language issues of both Italian and international students on both years of second cycle degree courses 1 . While she found differences between Italian and international students as regards evaluation of lecturer language and their own competence, with the latter being harsher in self-evaluation and less critical of NNS lecturers' language competence, she also noted a difference between firstyear and second-year students, with the latter being more tolerant of lecturers' English and having fewer problems in comprehension, possibly because their greater experience with EMI made them "able to reflect on the idea that language use is not just a question of the linguistic capacity of one or both parties involved, but an interaction between parties" (Clark, 2017, p. 303).
A specific comparison between the two cycles is particularly relevant for local students embarking on tertiary education in Italy, where students' English language competence, traditionally weak 2 , may have improved since the enforcement in 2014 of the Gelmini Reform (240/2010), obliging Italian secondary schools to offer some English-taught courses following the Content and Language Integrated Learning (CLIL) approach. This would imply that Italian undergraduates are now embarking on university education with greater familiarity with English-taught lessons and possibly better language skills than in the past. However little research has been done to evaluate the effects of this reform.
Our paper aims to provide insights into students' views of lecturer discourse by comparing the experiences of local first-and second-cycle students attending EMI courses at the University of Bologna (UNIBO). The experiences of international students attending the same courses are included as a benchmark for comparison.
Data for this study were collected via an online semi-structured survey of 128 local and international students attending undergraduate and postgraduate courses. Students were asked about their own English language competence as well as their impression of their Italian lecturers' language competence in English as to what extent this affects the lecturer's intelligibility in the classroom, particularly as regards speech rate, accent, and pronunciation.
Learners' opinions of their EMI lecturers' language competence as well as of their own linguistic ability in English are not the only variables likely to affect learners'perceptions of the degree of intelligibility in the EMI classroom. Kamaşak et al. (2021) also found that students with previous experience of EMI found lecture comprehension easier than those who had none. Local Italian L1 students in general might also find comprehension easier since they share the same NNS English variety with the lecturer (Fraser, 2006).
Students' responses in our study were investigated and compared in relation to their degree cycle: first or second. We also distinguish between Italian L1 and international students.

Methodology
Lecturers at UNIBO teaching first-and second-cycle courses in English in Economics or Engineering were contacted as part of a wider project regarding lecture discourse (Johnson & Picciuolo, 2022. Permission was asked to contact their students for the purpose of gathering further information. Students choosing to participate completed the survey via Google Forms. Participants were asked to select responses from a preset list, express their agreement on a Likert scale, or give a brief written response. Table 1 gives information about the students according to degree cycle and origin. We note slightly more first-cycle students overall (56%) than second-cycle (44%). While the L1 of all local students was Italian, a variety of languages was declared by the international students, with 65% of these belonging to one of three broad language sub-groups: Germanic
Students were asked to what extent lecturers' speed of delivery and pronunciation in general interfered with lecture comprehension, and how students coped in such cases. Responses are compared and discussed on the basis of degree cycle, as well as in relation to L1.

Students' English language competence
While the literature is divided as to what constitutes the ideal requirement for benefitting from courses in EMI (see Hultgren et al., 2022), providing documentation of English language skills at a >B2 level is still a pre-requisite for admission in many European universities. While most Italian L1 undergraduates declared a higher level than this (>C1), most Italian L1 second-cycle students declared >B2. This suggests that the younger Italian undergraduates have been able to benefit from enhanced English teaching including CLIL since enforcement of the Gelmini Reform, thus raising their own language competence closer to international language levels. By way of comparison, a greater proportion of international students (63%) declared a higher overall language competence (>C1) than Italian L1 students in general (53%).

Perceptions of lecturer's English language competence
When asked to specify where EMI lecturers should improve their language competence in order to perform effectively in lectures, both Italian L1s and international students indicated pronunciation and oral communication in general, in line with Costa (2017).

Undergraduates
Despite their criticism however, only 12% of local undergraduates had difficulty understanding, though this figure rose to 19% in the case of international students. Some Italian undergraduates reported initial difficulties in comprehension but noted that comprehension improved over time as they became accustomed to the lecturer's speaking style.
Most Italian L1 respondents and international students claimed that speed of delivery did not hinder comprehension, in line with Ackerley's (2017) findings. We suggest that problems are exacerbated the further the students' L1 language family from Italian, with Bangla, Serbian, Bulgarian and Tamil being the L1 of those international students who declared difficulties. We note also a possible lack of tolerance towards NNS on the part of the US native English student, who also claimed that fast speech and distinctive accent impeded comprehension. Other students mentioned the NNS accent as being problematic, drawing on their experience with NS accents:

Italian's English accent is different from US and understanding it difficult [Persian speaker]
some Profesors using Italian accent when they teach in the class make me difficult to understand, as I used to study in english-american accent during my previous degree.
[Bahasa Indonesian speaker] Though, as Jenkins (2014) notes, the benefits of a NNS rather than NS English-speaking lecturer are also understood:

Because it doesn't seem different from that which we speak back in my country (unlike in US or UK) [French/English speaker from Cameroon]
In this regard, Derwing and Munro (2015) point out that if the learner is familiar with the accent, s/he will also claim it is more intelligible. This suggests that the degree of accent familiarity might not only be related to the shared L1 of lecturer and learners, but, rather, to the degree of exposure students have had to varieties of English and particular accents.
82% of the international undergraduates said unclear pronunciation interfered with their understanding, with only the Spanish and Portuguese L1 speakers claiming not to have problems. Instead 58% of Italian L1 respondents said unclear pronunciation interfered with their understanding. These results suggest that students who share the same L1 as the lecturer or whose L1 is close to that of the lecturer have an advantage in understanding their lecturers' pronunciation (Bent & Bradlow, 2003).
As regards strategies for coping with pronunciation-related difficulties, most Italian L1 undergraduates would prefer to guess from the context than ask the lecturer directly, suggesting unwillingness to put themselves in the spotlight.
Students were also asked whether they would prefer the lecturer to use more simple language. Past research has shown that NNS lecturers tend to slow down their speech rate and use simpler sentence structures (Flowerdew & Miller, 2005), and this appeared sufficient for the Italian L1 undergraduates, with only 6% wishing for any further simplification. However, 30% of international undergraduates would prefer further simplification. This perhaps depends on the student's own level of English competence, since it was those with a lower level of English who wanted more simplification. However further simplification was also required by the US English native speaker, suggesting that other factors are at play.

Second-cycle students
83% of Italian L1 second-cycle students had no difficulty understanding lecturers' pronunciation, and indeed, whatever their L1, these students generally tended to have fewer problems with lecturers' pronunciation. This would suggest that maturity plays a more important role in students' lecture comprehension in the EMI classroom than shared language or even language competence. Just one international second-cycle student (Pakistan/ English home language) claimed pronunciation was not clear. This finding is in line with previous research by Clark (2017) who found that international students tend to be less critical as regards the quality of their NNS lecturers' English.
As regards strategies for coping with pronunciation-related issues, like undergraduates, Italian second-cycle students were also reluctant to ask the lecturer for clarification but prefer other means. The responses of most international students however show they are more likely to ask their classmates or even the lecturer directly in case of pronunciation problems. This suggests greater confidence due to greater maturity or experience.
40% of international second-cycle students would prefer the lecturer to use more simple language. Most of these students could not rely on similarities between their L1 and Italian to help them, and they also declared lower English competence. It should be remembered in fact that "EMI learners may experience greater language-related challenges according to their English level" (Aiwaza et al., 2020, p. 6), and indeed even a higher level of English proficiency "does not necessarily alleviate all of the challenges that students encounter" (Aiwaza et al., 2020, p. 16).

Would students prefer the lecturer to switch to L1?
Local students are still often the majority in international courses at UNIBO, but only 10% of Italian undergraduates and 21% of Italian second-cycle students would prefer the lecturer to switch to the shared L1 in these cases. The higher percentage in the latter case could be due to the lower language competence of second-cycle students.

Discussion and conclusions
Both Italian and international students said their EMI lecturers needed to improve pronunciation. Nevertheless nearly all Italian and most international students claimed that lecturer pronunciation and accent did not interfere with understanding, suggesting that it could be learners' familiarity with the lecturers' speaking style -in terms of speech rate, accented speech and pronunciation -that affects students' perception of their lecturers' language intelligibility. This contrasts with EMI lecturers' perception of their own language skills , where pronunciation is described as one of their major worries.
Our findings have shown clear distinction between cycles as regards language proficiency, and coping strategies for pronunciation-related issues. As regards language proficiency, Italian undergraduates overall had higher levels than Italian second-cycle students, thus confirming our hypothesis that the Gelmini Reform had led to better language skills among the younger Italian cohort. In comparison, international students reported a higher English language competence than the Italian students in general. As regards coping strategies, postgraduate students are more willing than undergraduates to ask either their peers or the lecturer directly for clarification. This could be due to their relative maturity when compared to undergraduates, as well as suggesting they feel more at ease with the learning environment of the EMI university classroom. Instead, students' L1 seems to have more influence than degree cycle as regards the desire for simplification, with those international students who could not rely on similarities between their own L1 and the EMI lecturer's L1 requiring further simplification. Second-cycle students in general have fewer problems than undergraduates with understanding due to pronunciation, speed of delivery and accent. As regards undergraduates however, internationals who did not share the lecturer's L1 or did not speak a language close to Italian had greater problem understanding than the locals. Overall, the majority of local students from both cycles would not switch to their L1 suggesting that they appreciate the added value of EMI.
Findings from this study have several pedagogical implications. Prospective undergraduate students, whatever their L1, would benefit more than second-cycle students from a content preparation course to gain familiarity with the English spoken by the local lecturers, while introducing them to their field of study and encouraging them to interact. Preparation courses for both students and lecturers (who also need to gain familiarity with student speech) could therefore also include audio-visual materials delivered in both synchronous and asynchronous formats.
Indeed, although much research has focused on training lecturers to improve their communicative skills in EMI settings, it is worth recalling that students are active participants in learning, and that any action taken in relation to improve comprehension in the EMI classroom should involve students and lecturers alike.