Establishing Teacher-Student Rapport in an English-Medium Instruction Class

English-medium instruction classes in higher education are increasing in countries where English is not the first language. Though these courses offer advantages, they also offer concerns and challenges. One of these challenges is creating a rapport between a teacher and students who are working in a language that is not their own. Rapport is important because it has been linked to academic performance. This study explored possible differences in rapport between two groups of students who were taking a class on Communications Research at a Mexican university. One group took the class in their native language, Spanish, and the other one took the class in English. The teacher, the programme, and the materials were identical in both cases; they only differed in the language of instruction. The study is qualitative in the ethnographic tradition. Videos of classroom interactions were used as a data source, which were analysed and coded. Codes were based on rapport-building activities. The teacher’s log and the nine-item Student-Instructor Rapport scale complemented the study. The videos showed evidence of rapport-building activities in both the Spanish and English groups. The results of the scale showed that the students in the English group perceived a stronger rapport between the instructor and the students, than those in the Spanish group.


Introduction
English is being increasingly used as the medium of instruction in universities from non-English speaking countries (Dearden, 2014), as using English to teach content classes allows universities to offer courses for international students. The benefits are both economic, as the university has a greater pool of candidates to choose from, and cultural, as both exchange and domestic students benefit from contact with people from other countries (Wächter & Maiworm, 2014).
In English-Medium Instruction (EMI) classes, the academic course content is offered in English, rather than in the country's language.
Teachers on these programmes are occasionally native English speakers and experts in the field of study. Sometimes, they are non-natives of the country, and English is the only common language they share with their students. More frequently, however, they are content experts, native to the country in which they teach, and non-native English speakers who offer their content classes in English as required by their university (Lara-Herrera, Richter, Razo-Colunga, & González-Espejel, 2016). This paper describes a study that took place at a small, private university in western Mexico. The university has exchange agreements with 145 universities around the world, welcoming approximately 50 exchange students every term, especially from Europe, but with some from Asia or North America, as well. The university offers around 50 content courses in English per term. Some of these, such as International Negotiations, cater especially to the exchange students, although Mexican students can also enrol in them. Others, such as Communications Research, are designed for local students, though they sometimes take in international students.
Most of the teachers in these courses are native Spanish speakers who have done master's or doctoral studies in English-speaking countries. Thus, they have a knowledge of academic English and of the terminology specific to their field. The university supports them by offering workshops about teaching content in English either on-site or abroad. The workshops include both language support and practice in adapting their materials and teaching styles for EMI.

UNIVERSIDAD DE LA SABANA DEPARTMENT OF FOREIGN LANGUAGES AND CULTURES
During a reflection activity after one such on-site workshop, teachers were asked about the challenges they expected to face while teaching their classes in English. A few instructors worried that the students would speak the language better than they did. Some teachers mentioned having to restructure their syllabus and find new materials.
However, one issue that came up repeatedly was the importance of the teacher-student relationship. One professor said, "I don't feel I'm myself when I speak English" (MR, personal communication). Another teacher shared, "I'm not as funny in English" (AS, personal communication). A third teacher reported, "I find it hard to establish rapport with my students when the class is in English" (FS, personal communication). These three professors all had at least 10 years' experience teaching at the tertiary level.
Their concern was not lack of pedagogical or content knowledge.
It was not even a lack of linguistic skills in English. Rather, their concern was with the loss of a skill that adds much to the learning experience: that is, the ability to create a personal connection between students and teacher. This personal connection is usually created through language: small talk, humour, or instances of personal disclosure to create a bond between interlocutors. These teachers were concerned that not using their own language would make it difficult for them to establish a rapport with their students.
The university is small -with about 3,500 undergraduate students -, and one of its educational goals is to offer personalized attention to its students. Classes tend to be small, and teachers usually know students by name, or even by their nicknames. Thus, feeling "not oneself," or being unable to establish rapport with the students, are important issues in this context. The concerns expressed gave rise to a few questions: Does the language of instruction create a barrier to rapport between teacher and students? Or rather, is it the fact that teacher and students are communicating in a language that is not their own?
And of course, do the benefits of EMI outweigh the loss of rapport?
To answer the questions, this study turned to a bilingual professor of Communications Research. She taught two sections of the same course: one in English, and the other in Spanish, and these two courses were compared in terms of the rapport created between instructor and learners.

English as a medium of instruction
The 21 st century has seen the expansion of English-medium instruction in higher education institutions throughout the world. EMI allows an institution to become more international, attracting both students and instructors from around the globe. This can be especially advantageous to institutions where the national language is not commonly spoken outside the country, such as Iceland, Sweden, Denmark, and Finland (Wächter & Maiworm, 2014).
Implementing EMI has been easier in some contexts than in others, however. Dimova, Hultgren, and Jensen (2015) mention that, in some countries, such as Italy, EMI has been viewed as a potential threat to the national language, to teacher freedom, and to student learning. Dearden (2014) reports that Indonesia reversed its policy on EMI instruction for fear it would "endanger the national identity" (p.19), with English becoming the language of the elites. Even in northern Europe, where implementation of EMI has been relatively smooth, Airey (2015) has found important concerns among instructors on these courses, including the need for more preparation, lack of training, less fluency, less flexibility, and the need to speak more slowly, thus covering less material and in less detail throughout the course. Airey (2015), who has carried out significant research into EMI in Sweden, emphasizes the need to consider the discipline the course belongs to. For example, he states that natural science courses seem to lend themselves better to EMI than those of social sciences. It is important to consider, however, that students in the fields of social sciences tend to have greater English language proficiency than those of engineering or sciences (Wächter & Maiworm, 2014). EMI courses in Europe increased from 725 programmes in 2001, to over 8,000 in 2014 (Wächter & Maiworm, 2014 Pun has looked into EMI in Asia.

UNIVERSIDAD DE LA SABANA DEPARTMENT OF FOREIGN LANGUAGES AND CULTURES
Little research on the topic has been carried out in Africa or Latin America, so it is difficult to say how it is being implemented in those regions. Though EMI is growing in Latin America, the only Latin American study found at the time of this writing is from Colombia (Corrales, Paba-Rey, & Escamilla-Santiago, 2016). Both Dearden's (2014) overview of EMI as a global phenomenon and Macaro, Curle, Pun, An, and Dearden's (2018) article on the state of the art in EMI in tertiary education include only one Latin American country: Colombia.
A summary of these studies would indicate that both instructors and students accept that their English language skills may not be perfect, but that they can work together to make sure the content is clear (Ball & Lindsay, 2013). Teachers tend to have compensatory strategies to get their ideas across. These include using humour and taking advantage of students' language knowledge to bridge the gap (Kling, 2015; 2016), as well as use of students' first language, especially for introducing new vocabulary (Macaro, 2019). In her study of teacher identity, Kling's (2015) participants -nine Danish instructors -described a good teacher as "knowledgeable, engaging, organized, interactive, memorable, and enthusiastic" (p. 230). Thus, teachers felt that passion for their subject was much more important than language skills. As for the students, essential competencies needed to learn successfully on an EMI course are motivation (Doiz & Lasagabaster, 2018), critical thinking skills, learner autonomy, creativity (Sancho-Guinda & Breeze, 2017), and self-efficacy (Thompson, Aizawa, Curle, & Rose, 2019).
Other studies that have compared classes taught in English or Spanish by one teacher include Maiz-Arévalo (2017) who looked specifically at the use of questions in interactive lectures and found that the teacher tended to ask more questions in English than in Spanish. Breeze and Dafouz (2017) compared student answers on a written test in courses taught in English and Spanish by one instructor and found the writing showed similar problems regardless of language of response. In both studies, the instructors were native speakers of Spanish with a good, but not native, level of English.

Rapport
Studies referenced in the previous section (Ball & Lindsay, 2013;Kling, 2015; for example) emphasize that learners worry less about the instructor's language proficiency, and more about his or her teaching skills.
Among the latter, the ability to establish rapport with students is vital. Frisby and Martin (2010) define rapport as "an overall feeling between two people encompassing a mutual, trusting, and prosocial bond" (p. 147). Rapport is mentioned by students as an essential characteristic of teachers, and it consistently predicts student engagement, as well as affective and cognitive learning (Frisby & Martin, 2010). Buskist, Sikorski, Buckley, and Saville (2002) state that the effective teacher is "approachable, genuine, humorous" (p. 28), as well as both respectful and respected, and that these characteristics help establish rapport in the classroom.
Rapport is developed through immediacy expressed by both verbal and non-verbal teacher behaviours. Using the students' names, sharing personal stories, using humour, making eye contact, smiling, nodding, and praising all contribute to immediacy, and thus, to building rapport (Benson, Cohen, & Buskist, 2005;Estepp & Roberts 2015;Wilson, Ryan, & Pugh, 2010). Though rapport seems to have a positive effect on student motivation and attitudes (Wilson & Ryan, 2013), few instructors regard it as important (Benson, Cohen, & Buskist, 2005).
Many studies explore the challenges of implementing EMI, yet few mention rapport building as one of the challenges. Among these, Dafouz-Milne (2011) acknowledges the importance of rapport in facilitating teacher-student interactions and recommends including training in social small talk as part of EMI teacher development. Ibrahim (2001) mentions teachers' code switching (i.e., reverting to their and the students' common language) in EMI classes in order to create rapport.
According to Sert (2008), 84% of the Turkish EMI instructors she interviewed mentioned that, the most important problem they experienced was that their lessons lacked a sense of humour because even though they spoke English fluently, they could not create colour and spirit in a foreign language. Thus, they complained that they gave their courses in an environment which was neither lively nor joyful. (p. 165) Kling (2015) found that experienced teachers compensated for their lack of language skills with their content expertise and with humour, thus precluding any potential difficulties. Jernigan (2017) discusses the importance of authenticity in the EMI class, not only of the materials, but of the instructor, as well. For her, authenticity includes self-confidence and a willingness to appear less than perfect before the students.
Studies into English as a second language classes show the importance of teacher small talk in creating rapport (Ayala-González, Leonel de Cervantes-Orozco, González-Cabrera, Romero-Mayoral, & Mugford-Fowler, 2011;Santana-Villegas, 2014). This last author mentions that it is especially important for Latin American students, who value "social cohesiveness and individual recognition" (p. 20).

Method
The course on Communications Research is offered at the university in this study to undergraduates in the fields of Journalism, Audio-visual Communications, or Public Relations and Advertising. It is a required course and is taken after the fourth semester of studies. It is usually offered in English because the teacher is bilingual, and it also adds to the course offerings for international students. Between one and five international students sign up for the course in any given semester.
In the winter term of 2019 (January to May), two sections of the course opened. For the purposes of this study, it was decided to offer one in Spanish and the other one in English, but both led by the same instructor, who is bilingual and feels comfortable teaching in either language.
The study focused on the research question: How does the language of instruction affect rapport between students and teacher?
The study is qualitative in the ethnographic tradition. Video recordings were made of the classes, and they were analysed and coded, looking for indicators of rapport. Ethnographic studies are useful in studying patterns of behaviour in groups because they allow data to be collected on day-to-day interactions over a length of time (Creswell, 2012). Ethnography is characterised by being carried out in the field; that is, where the studied group normally interacts, and not in an experimental situation. Observation and interviews are usually used for data collection, and these are analysed in search of cultural themes and common patterns of behaviour (Creswell, 2012).
Video ethnography, or the use of video to record interactions, is especially useful where the researcher is also one of the participants (Goldman, 2014), as in the case of this study. For Goldman (2014) work on the projects they have designed; they carry out the study and they present their findings and conclusions. Thus, the course is practical, rather than theoretical. Little time is spent on teacher lectures and more time is spent on hands-on activities where the students work either individually or collaboratively. In EMI contexts, practical courses may be more effective than lectures (Airey, 2015;Ball & Lindsay, 2013).
The latter authors found that students worry less about the language of the class and more about the pedagogy, and that they prefer practical to theoretical courses. In these cases, they are willing to accept that the instructor's English is not perfect if the class is well structured.
The Communications Research course offered at this university follows a Problem-Based Learning (PBL) approach (Peñalosa-Castro & Santana, 2018). Wilkinson (2013) has found that PBL courses may promote learning in EMI because students are active in using language (defining, describing, explaining, accounting for, differentiating, etc.); students are in charge of their own learning; problems can be pre-designed both to elicit content knowledge and to activate and develop language competences; collaborative learning allows mutual help during the PBL process; productive tasks in PBL allows assessment of content and language; and there is limited reliance on staff language ability. (pp.15-16) On the first day of class, students were given an explanation of the nature of the study. They were told the study focused on how the teacher interacted with the students and how the language of instruction affected the interaction. The word rapport was not specifically mentioned. The participants were told that the classes would be video recorded and analysed. They were given a consent form to sign if they agreed to participate in the study. They were told they could opt out if they did not want to participate, but none did. They were also given the option of changing class if they preferred to study in Spanish. Three students changed from the English to the Spanish section, and one changed from Spanish to English. Additionally, a French exchange student signed up for the course. Her data are not included in the study.

Participants
The Spanish class consisted of 34 students, 29 female and five male. The English class consisted of 30 students, 25 female and five male; the age range in the two groups was 19 to 23, and the average age was 20.23 years old.
Participants were asked about their level of English. Of the 64 participants, four from the Spanish group reported their level as Intermediate 2, according to the levels offered at the university. This is equivalent to the B1 level described by the Common European Framework of Reference (CEFR). The other 60 participants reported their level of English as B2 or better, according to the CEFR.
The instructor is also the researcher on this study. She has been teaching for over 30 years, so she is much older than the students are.
She holds a first degree in Communication Sciences, a master's in Pedagogy, and a doctorate in Education, as well as a diploma in teaching English as a foreign language. Thus, she has knowledge of content, pedagogy, and language, but she also has experience in gauging how much language a non-native speaker can handle, and how to rephrase to make meaning clearer.

Data collection
Data sources were video recordings of class sessions, the teacher's daily log of activities, and a questionnaire applied at the end of the course.

Video recordings
A video camera was set up on a tripod on the teacher's desk at the front of the classroom. From that angle, most of the classroom and the participants were visible. Only the top right-hand corner of the classroom was not visible, so the students sitting in the seats located there did not appear on the videos. The sound quality was tested and both teacher and students were audible. Not all of the sessions were recorded.
Included were classes where the professor interacted with the whole group. Exams, student presentations, or classes where the professor worked with students on their individual research projects were not included. Because of this, the eight recordings from each group are from the earlier part of the course, especially the first sessions.

UNIVERSIDAD DE LA SABANA DEPARTMENT OF FOREIGN LANGUAGES AND CULTURES
The teacher did not view the recordings until the end of the course, to avoid a washback effect. It is possible that knowing that rapport was the issue under study did have an impact, with the instructor making a concerted effort to develop rapport between herself and the learners.
Once the course had finished, the videos were coded according to previously established categories, which will be detailed below.

Teacher's log
As each class finished, the teacher recorded in her log her impressions of the class. She took note of how students had responded to the activities, what had been especially difficult or too easy, where she had run out of time, or where she had run out of activities. Based on these reflections, she adjusted the class. For the purposes of this research, the teacher's class log served to contrast the responses of the two groups.

Student-Instructor Rapport Scale
At the end of the term, the participants responded to a questionnaire on their perceptions of rapport in the class. This questionnaire was based on the Student-Instructor Rapport Scale (SIRS-9) (Lammers & Gillaspy, 2013). It also served as the basis for the coding categories, as will be explained in the following section.

Data analysis
The videos were uploaded to a cloud site and viewed and coded independently by three different researchers (the instructor and another two persons who were not involved in teaching the course).
The unit of analysis was the episode, which was defined as a brief period of interaction between instructor and learners. An episode was assigned to a category if at least two of the viewers had coded it at such.
The camera was switched on before the class began and switched off after the students had left. Thus, each recording lasted between 80 and 90 minutes.

Coding categories
Lammers developed a nine-item Student-Instructor Rapport Scale for use with university students (Lammers & Gillaspy, 2013). He mentions that, though rapport between students and teachers is important to 9. In general, you are satisfied with your relationship with the instructor. (Lammers & Gillaspy, 2013) Information about how the scale was validated can be found in Lammers and Gillaspy (2013). These items were used to create coding categories for the analysis of the videos of teacher/student behaviours in class. Thus, the following categories were developed, in which the instructor: • makes an effort to understand students • makes an effort to encourage students • cares about the students • treats the students fairly • communicates effectively with the students • respects the students • gains the students' respect • is approachable when students have questions.
Two additional categories were created, based on concerns expressed by other EMI instructors at the same university, and in accordance with Buskist, et al. (2002), in which the instructor: • uses humour • is authentic

Results
Analysis of the videos shows that the classroom is an auditorium-like setting and the teacher moves frequently to the back of the classroom on one side and another. She uses the board infrequently, usually to write the objectives of the day's class or to make note of something. A laptop computer is connected to a projector, and most of the class material is projected. She uses presentations, videos, or documents uploaded either to a learning management system or to a shared cloud site.
The environment tends to be rather informal. Students walk into or out of the class at will and they spend a lot of time chatting. This is a characteristic of the School of Communication, but not so much of the university in general.
Many of the students refer to the teacher by her first name; a few call her "teacher" which is a translation of the respectful Maestra with which many Spanish-speaking students address their instructors. In the Spanish class, approximately half the group use the informal tú to address her; this is also common in the School of Communication, where even the dean is known by his first name. The other half uses the more formal usted form, possibly in deference to the teacher's age.
The teacher always refers to the students by their first names or by their nicknames.
The teacher usually arrives to class a few minutes early and spends some time chatting informally with the students as she sets up the laptop and the materials. These conversations are always in English, regardless of the class; she switches to Spanish to signal the beginning of the session in the Spanish group. In the English group, she says, "Let's begin," or "Let's get started," or a similar phrase.
In the Spanish group, on two occasions, two different students ask a question in English, possibly because the material is in English. In the English group, one student consistently addresses the teacher in Spanish. With the English group, the professor uses only English, though she occasionally translates a term or she asks, "How do you say this in Spanish?" After the class, if the students approach her with a question, she will answer in the language the students use; that is, if the student asks in English, she responds in English. If the student asks in Spanish, she responds in Spanish.

PP. 265-291
A typical example would be the first session. The instructor goes over the syllabus and how the course is divided. She explains the rules of the class, and then introduces a short activity. She shows the students a popular meme that decries the linguistic poverty of young people today. To determine if the meme is true, the students work in small groups to content analyse the lyrics to different regaettón songs, then they present their conclusions. In the English group, one of the female students says of the lyrics, "This is horrible!" The teacher replies: "Day 1: L discovers that regaettón is vulgar". The students laugh. Despite this touch of humour, the videos of those first session shows that the teacher is chattier with the Spanish group than with the English group. This may be because she still is not sure how much they can understand in English. The video of this first session of the Spanish class shows numerous instances of the instructor approaching groups of students as they work. In the video of the English class, the instructor appears more distant. She remains at the front of the classroom for most of the activity. Subsequent videos do not show this difference.
The teacher's log for this session says, Differences I noticed: I'm less chatty in English. Students had fewer spontaneous comments. Is this an individual difference, or related to the language? (Teacher's log, January 14)

Making an effort to understand the students
The teacher tends to rephrase students' questions or comments in either language to make sure she has understood before attempting to answer. She will say, for example, "Do you mean…?", and rephrase what the student has asked. However, there are two students -one in each group -who appear to be totally disengaged. They do not participate spontaneously; they never ask questions nor make comments. One of them spends a lot of time on her cell phone, yet the teacher does not attempt to engage her or to understand why she does not participate.

Encouraging the students and caring
The teacher does encourage the students. On three occasions, the videos show her talking to individual students as she checks their work.

UNIVERSIDAD DE LA SABANA DEPARTMENT OF FOREIGN LANGUAGES AND CULTURES
"Oh, that's so beautiful," she says to one, and "Wow, this is so great!" to another student in the English group. To one of the students in the Spanish group she says, "Great job! That's excellent".
One of the students was hospitalised and missed an exam. When he arrives to class after this, she asks him how he is doing. After class, he approaches to ask what will happen with the grade. The teacher explains that she will count his coursework as an exam. This episode was coded by the researchers as evidence of caring.
On the last day of class, the teacher bids farewell to the students.
She thanks them for their participation and wishes them good luck on their exams and a happy vacation. Three of the students from the Spanish class hug the teacher as they leave the classroom and thank her for the class. Several students from the English group also thank her, but there are no hugs.

Communicating effectively
In another session, the topic is the differences among qualitative, quantitative and mixed methods research. Students are divided into three teams, with one captain per team. Each team is assigned one type of research. They look for definitions and examples of use. Then, they work in groups of three and complete a compare-contrast chart with information about the three types of research.
In the Spanish group, there are more questions before they begin working, and the team captains are more controlling. The teacher asks the students to colour-code the work to show differences and similarities. They have trouble understanding these instructions and begin to show frustration. "Do you mean, like a Venn diagram?" one student asks. "Yes," the teacher answers, "but I couldn't make a Venn diagram with the computer". The student laughs and draws a Venn diagram by hand on a sheet of paper. The teacher simplifies the instructions and the learners complete the work. In the English session, she does not ask the students to colour code; she has adjusted instructions from the Spanish class. Communicating effectively, especially instructions, is an area the professor needs to work on, as evidenced by this -and other -episodes.
The teacher's log gives more information: In this session in particular, it is interesting to note that, in the English group, there are fewer questions, and the captains are less controlling. However, the Spanish work is better. As the teacher's log notes, in the English class, the participants merely cut and pasted the information from the internet onto their charts. In the Spanish group, the video shows the students discussing the information to understand and summarize it. It is difficult to say if these differences are due to the language of instruction, or the characteristics of the members of each group. In general, though, this trend can be seen in other sessions.
The Spanish group asks more and deeper questions, while the English group gets to work more quickly, but their work is more superficial.
The teacher's log shows another instance from student presentations, which were not recorded. The presentations dealt with different communication theories. This is material the learners had seen in their first semester of classes and these presentations were intended as a reminder. Students were divided into six teams. They were asked to find information about six different communication theories and present it to their classmates. The teacher's log states, I notice that presentations in Spanish are better, longer, more complete. Also, they are presented and not merely read off the slides. Note: everyone in Sp group had references [i.e., they cited where they had gotten their information], whereas only one team in Eng group did. Does this mean that Sp group is just better than the other group? (Teacher's log; Jan. 21)

UNIVERSIDAD DE LA SABANA DEPARTMENT OF FOREIGN LANGUAGES AND CULTURES
One notable difference, however, was in the research projects themselves. In one session, the participants were asked to think of a project they would like to work on. There were three requirements: it had to be in the field of communications; it had to be doable in the time and with the resources they had, and it had to be something they were personally interested in. In the following session, the students formed round tables and explained their projects to their peers. The idea was to "sell" their ideas and find someone they wanted to work with. In the Spanish group, the choices were based on the project: "That's interesting. I want to work on that," whereas in the English group, the choice was based on the person: "I want to work with S". However, the projects in the English classes tended to be more varied and complex than those of the Spanish group.
The teacher's log shows: Students decide on a research problem and think possible population, collection strategies and paradigm. Spanish group spends a lot of time on this, but English group spends less. However, English group projects seem a little better (Teacher's log, Feb. 6).

Being approachable
In particular, two students in the Spanish group tend to ask deep questions. On several occasions, the videos show the instructor approaching the student who asks a question and then the teacher spends a few minutes talking individually with her, while the others are doing something else. This was coded as being approachable. Additionally, on several occasions, the videos show the students approaching the teacher after the class has been dismissed to ask questions or clarify doubts.
These instances were coded in the same way. This is more common among the students from the Spanish group than among the English group; thus, the questions are not related to language issues.

Being respectful and respected
Students tend to be chatty and distracted. The instructor believes these are characteristics of students in this particular field of study and in this generation. She does not equate distraction or chattiness with lack of respect. There are three especially chatty students in the English group. When they interrupt the flow of the class, the instructor stops talking and waits for them to pay attention. They do so with a smile and an apology. These episodes were coded as both respect for the student and for the teacher.
As to the use of the formal usted among the Spanish group, it is difficult to reach a conclusion. It does signal respect, but not using it signals connectedness (for example, a child does not normally use usted with its grandmother, regardless of the age difference). Both respect and connectedness could be indicators of rapport.
It is difficult to find examples of fair or unfair treatment in the videos; none were coded as such.

Being authentic
The instructor is a soccer fan. The city where the study was carried out has two professional teams in the first division. The teacher and two of the students in the Spanish group are fans of one of these teams, while another student is a fan of the other. The videos show some banter before the class as they talk about the results of the weekend matches.
These were coded as the teacher being "herself" (i.e., authentic). In the same category is an episode where the teacher tells a short anecdote to illustrate a point. She is explaining what constitutes evidence, and she talks about an incident with an ex-boyfriend. In both groups, the students lean forward, they laugh, they ask for more details.

Using humour
In instances of humour, the instructor sometimes uses memes to introduce or illustrate topics in light-hearted manner. For example, to explain the difference between general and specific objectives, she uses a popular song. She sometimes shows another meme of an image of a saint, looking towards heaven. It says, "When you've just given instructions for the third time, and a student asks 'What are we going to do?'." The students invariably laugh, but it serves as a reminder to pay attention to instructions.

UNIVERSIDAD DE LA SABANA DEPARTMENT OF FOREIGN LANGUAGES AND CULTURES
The instructor never uses jokes overtly, but the videos show multiple episodes of student laughter. In general, the class atmosphere with both groups is relaxed.

Rapport scale
The students responded to the SIRS-9 (Lammers & Gillaspy, 2013) at the end of the semester. The scale was translated into Spanish and delivered to the participants via a Google Form uploaded to the learning management system. They were asked to respond, but there was no incentive to do so; 40 students out of the 64 responded (62.5% of the total number of students); 22 were from the Spanish group and 18 were from the English group. Table 1 shows how the students responded to the questionnaire.
Results are expressed in mean points out of five possible points. As can be seen in Table 1, both groups of students were satisfied with the rapport developed between themselves and their instructor during the course. The students in the English course were even more satisfied than those of the Spanish course. A last question on the questionnaire asked, "Is there anything else you want to say?" One of the members of the Spanish group complained the activities were tedious. The rest of the responses were positive. They mentioned the teacher was knowledgeable and patient. They felt the activities were dynamic. One participant mentioned that the material learned on the course would be useful for the future, and one student mentioned that the fact that the class was in English did not make it more difficult. Another student stated that having the class in that language was beneficial, "I feel my mind works better and I remember things more because I have to be concentrated, not only because of the material, but because of the language itself" (anonymous student response). These responses are in line with Ball and Lindsay's (2013) findings that students are more concerned with the class pedagogy than with the language of instruction.

Discussion
English-medium instruction classes in higher education are increasing in English non-dominant countries around the world. Though these courses offer advantages, they also offer concerns and challenges. One of these challenges is creating rapport between a teacher and students who are working in a language that is not their own. This study sought possible differences in rapport between two groups of students who were taking a class on Communications Research. One group took the class in their native Spanish; the other group took the class in English.
The teacher, the program, and the materials were identical in both cases; they only differed in language of instruction.
Establishing rapport between the teacher and the students is important, as it can lead to greater engagement, motivation, and better learning outcomes (Buskist, Sikorski, Buckley, & Saville, 2002;Frisby & Martin, 2010). However, as student responses on the questionnaire and the evidence of the videos show, the language of instruction per se is not a barrier to rapport. The videos show how the instructor creates immediacy, and thus, rapport, using humour, praise, anecdotes, and self-deprecation (as evidenced by the episode of the Venn diagram) UNIVERSIDAD DE LA SABANA DEPARTMENT OF FOREIGN LANGUAGES AND CULTURES (Benson, Cohen, & Buskist, 2005;Estepp & Roberts, 2015;Wilson, Ryan, & Pugh 2010). She establishes herself as an authentic and caring person, rather than as a perfect teacher.
This study shows that the language used in class is not a barrier to immediacy. If rapport fails to develop in an EMI course, other factors, such as lack of linguistic ability of either teachers or students, or teaching style (i.e., reliance on lectures) may be the cause (Dafouz, 2011;Sert, 2008). This paper in no way intends to imply that only native speaker English teachers should give EMI classes. Rather, it is believed that a student-centred class, such as PBL, in this case, lends itself better to developing rapport between students and teacher than a lecture-based approach (Wilkinson, 2013). Additionally, the study serves to underscore Dafouz-Milne's (2011) suggestion to include training in small talk in further EMI workshops.

Limitations
It is, of course, impossible to generalize from the results of such a small-scale study. It is also necessary to consider that placing a camera at the front of a classroom may have affected the behaviour of both learners and instructor. While observing only one teacher can also be considered a limitation, in the case, it helped to show that language of instruction per se is not an obstacle to rapport building. Having the instructor take on a role as lead researcher is another source of concern, though efforts were made to reduce this bias. Among these efforts were having multiple coders and not viewing the videos until the study had concluded.
Thus, in spite of its limitations, the fact that both courses in this study were identical, except for the language of instruction and the students themselves, allows this research to make a contribution to the literature on English-medium instruction in higher education.