How Can Emotion-full Language Affect The Teacher/Student Relationship?: Evidence From Intercultural EFL Online Chat Tutoring



As the world becomes more interconnected, new methods in the study and practice of language learning are needed to account for the experience of a globalized world (Bloomaert, 2010) and the continual intertwining of technology into our lives at younger and younger ages (Pasfield-Neofitou, 2013). Simultaneously, teachers need to know not just the new pedagogy, but general CMC principles related to social roles and importantly, emotion. This study seeks to investigate how emotion is expressed and understood in intercultural, Korean student (KE), non-Korean teacher (NE), English language tutoring through the synchronous chat program, Google chat.

As a teacher and participant in this study, the experience of tutoring a new student, with new pedagogical approaches and theoretical designs, what was most impressive to me was the way I, as a teacher, expressed and received emotion. Comfortable chat conventions that are normally used in my day-to-day chat experience were not always or consistently found in the tutoring sessions. Understanding why, particularly if it was due to some lack of skill on my part, became an important question.

An ecological perspective (Gibson, 1977) would suggest that computer-mediated communication (CMC) would afford different types of interactions between teachers and students in terms of content, style or conventions. Recent studies (see Derks et al, 2007 for a review) have found that emotional cues in CMC environments which lack many of the visual and non-verbal cues are reinterpreted for computer and writing environments. However, while many researchers have investigated asynchronous CMC (ACMC) environments like email, relatively little has been said about synchronous CMC (SCMC) such as chat programs. Even less has been written about the teacher/student relationship (Bretag, 2006). This study attempts to investigate two questions:

  1. How do teachers and students express emotion and does emotion communication improve intercultural competence (ICC) in SCMC?
  2. In what ways do the SCMC affordances impact ICC?

This study will be presented in the following format. First, a brief review of the relevant theory and recent studies related to emotion in ICC and CMC environments. The present study will be described including who the participants are and how the data was collected and analyzed. The results as they pertain to the research questions will be presented, followed by a discussion of the implications of the results for current theory and English-teaching pedagogy. The study will be concluded with the limitations of the method and results and directions for future research will be presented.

Theoretical Review

2.1 Emotion in ICC

Since Byram’s (1997) influential model of competent intercultural communication, others have built upon that foundation and expanded the role of emotion. Matsumoto et al (2007) have described a model of intercultural communication and competence as one that is predicated upon emotional regulation. In this sense, they take the attitude portion of Byram’s (1997) model and privilege it above the other categories as a psychological gatekeeper. Communicative content is first filtered through our emotions before it can be open to interpretation and interaction. How an individual reacts to the emotional content of an IC encounter can determine if specific instances are exaggerated or missed altogether.

Gullekson and Tucker (2014) studied how emotional regulation, what they described as “emotional intelligence” (p. 3) affected American university students who did short-term (a month or less) study abroad in various countries. They found that scores of emotional intelligence predicted levels of satisfaction and perceived cultural awareness in the students. They suggest that students preparing for study-abroad should be educated not just in cultural awareness, but that they be prepared emotionally as well, suggesting that emotional communication is an important aspect of navigating intercultural situations.

2.2 Emotion in CMC

While Matsumoto et al’s (2007) model was in part an attempt to capture both the real life face-to-face (f2f) and CMC interactions related to emotion, Derks et al (2007) examines how emotion is expressed specifically in CMC environments. They found that online content is filtered through very similar categories as f2f interactions, but in different ways. Derks et al (2007) use the concepts of the physical and social, crossed with implicit and explicit.


(other non-verbal cues)


(norms and expectations)


(emotional labels)

^^ 😦 🙂 XD haha ㅋㅋㅋ … !!! “Good! But maybe…”

(personal involvement)

Quick response time “anyways…”
Table 1: Adapted from Derks et al (2007) table of emotion categories in    CMC.

communication styles to categorize emotion communication in CMC (see table ). Understood together with the ecological perspective of Lamy and Hampel (2007) it is possible to create a “semiotic budget” (van Lier, 2000) for any CMC environment.

Specifically, Derks et al (2007) define emotion communication as the expression, understanding and sharing of emotions. They note that previous research has suggested that the lack of actual physicality in CMC gives it a certain anonymity and exaggerated style. For example, men were found to be more negative and critical in online forums when compared to f2f interactions (Castellá et al, 2000). However more recently, researchers have found that physical emotional cues are expressed different online, and that the situated nature of the CMC interaction can regulate the expressions related to anonymity (Scissors et al, 2009).

2.3 Communication Accommodation Theory

Giles’ (2008) communication accommodation theory (CAT) is one way to understand Derks et al’s (2007) sharing. CAT is centered around the idea of interpersonal convergence and divergence in communication. The theory is an attempt to explain why some people have positive communication experience at one time and bad ones at another. To do so, CAT considers how two or more interlocutors choose to accommodate, or change their communication, to each other. Convergence in CAT is related to similarity in speaking style, even if the two speakers don’t speak similar dialects. Divergence occurs when one or more of the interlocutors does not wish to create a sense of closeness towards the other or is trying to maintain group boundaries.

Scissors et al (2009) examined the concept of trust between people in chat-based SCMC environments using CAT. Participants were given a dilemma task and asked to solve the problem together using a chat feature. They found that, among other things, emotional content and emotional stylistics (e.g. lol) indicated higher levels of trust than pairs that did not include such communication.

However, less examined is the role of emotion in ICC and CMC between teacher and student. Bretag (2006) is one example where this attempted has been made in ACMC email environments. Bretag (2006) found that the teacher is not just a transmitter of culture, but creates a “third space” (p. 981) where both teacher and student can re-create identity and emote differently outside of their f2f interactions. However Bretag (2006) focused on understanding the types of communication styles used in these interactions, without a specific focus on emotional communication.

3. Method

3.1 Participants and site

Participants for this study come from a teacher-training tutoring experience where a class of TESOL master’s students tutored a group of undergraduate Korean students at a university in Seoul, South Korea. The tutors were both Korean and non-Korean and were paired with an undergraduate for three sessions of about one hour each sessions over three weeks. The purpose of the tutoring was unrelated to the present study; instead, the tutors were interested in developing instructional conversation skills (Meskill & Anthony, 2010) with task-based activities in SCMC environments.

Of the tutor/students in this training, six pairs were chosen for this study. Because we are interested in the inter-cultural aspects of SCMC, only foreign teachers were selected. Three males and three females from the United States and Canada were chosen along with their Korean students. In total, there were twelve participants altogether, including the researcher.

The tutoring sessions occurred using google chat, normally on monday nights unless the participants needed to schedule other times that worked for both of them, which happened a couple of times with the participants in this study. The google chat platform is similar to many chat platforms in that it allows chatters to use pre-made emoticons and incorporate the use of GIFs, or animated emoticons.

3.2 Data Collection

            3.2.1 Grounded approach

Chat logs from the six tutor/student pairs were analyzed as data for their emotional content and environmental affordances. While chat logs can present a rich array of data, including time stamps and linguistic enhancements like hyperlinks, each tutor copied and pasted their data into a wiki file and different forms emerged from that process. As such, certain information related to time-between-responses and certain emoticons like GIFs were not available.

This study utilized a qualitative grounded theory approach to data collection and analysis. A grounded approach relies on discovering themes and categories that emerge from the data and not from a constricting pre-conceived theory (Creswell, 2003). After the emotional categories present in the data emerge, the researcher used a conversational analysis journal (see appendix) to examine the emotional interactions of the participants.

            3.2.2 Emotional Categories

In order to answer our first question, the data was first organized by tutor/student pair and chronologically by session. The first session of each dyad was read carefully and marked for any emotional content that happened, such as use of emoticons, exclamations, hesitations, gratitude or encouragement. These categories (see table 3), were noted and the data was analyzed once again. The researcher annotated each individual instance of emotion with a specific hashtag, such as #GRT for “encouragement”.

Dyad Who initiated? Response? Turns? Closeness / distance?
1 Teacher







7 No – T tries to point out errors with praise, S hesitantly answers, T praises.
4 Teacher


(grt) 🙂 (!)



(grat) 🙂 (!)


4 Closeness – the pair takes their leave.
Table 2: Discourse analysis journal

Once the emotional categories were discovered they were searched for various styles and conventions that fit under one category, such as J or “turn-taking”, for encouragement. Next, they were examined for initiation-response interactions. In order to determine the importance of emotional language, instances of emotional language which elicit emotional responses are considered evidence that emotion played role in the closeness/distance of the tutor/student communication. Important for our purposes in this study, tutor or student responses to emotion communication constitute instances of interactional emotion communication (IEC). How the dyad responds, either with positivity, hesitancy or neutrality is used to define the emotional accommodation of the participants. The researcher highlighted each emotional interaction and used the analysis journal to determine what kind of emotions were expressed, what the reactions were, how many turns each interaction took and a brief summary of the content.

Two tutoring sessions were analyzed in this way before a comparison of the different emotional interactions was analyzed to determine how emotional communication changed between sessions.

3.2.3 Affordances of SCMC in ICC

Following the analysis of emotional categories, the varying affordances of SCMC in ICC interactions was determined by examining the personal reflections of the tutors about their experience in the interactions. While their personal reflections were not directed towards the topic of emotion, how they reacted to the SCMC medium provides evidence of positive or negative feelings. Additionally, the tutors reflected on how they could have performed better with their student. Any evidence of emotional knowledge was considered evidence of ICC awareness.

The researcher also reflected personally on their own experience of the tutoring session and how the experience might relate specifically to the concept of emotion in ICC. Following the framework created by Lamy and Hampel (2007), affordances were organized in terms of time and mode. These two categories were crossed with the emotional characteristics of CMC identified in Derks, et al (2007), namely physical and non-verbal, written language, and social norms (see table 6).


4.1 Emotional Categories

Emotional categories observed in the tutoring sessions are familiar to any teacher and are listed in table 3. Between sessions, the most common form of emotion communication was encouragement and this was predominately used by teacher towards students (18.7 uses per session vs 0.8). Teachers used positivity to enhance their EC by using emoticons and hyper-use of exclamation marks more often than the students did. Students were more likely to communicate uncertainty or hesitancy and to apologize.

  Average occurrence per session
laugh smile   encourage hesitate gratitude apologize negativity
teachers 2.7 4.0   18.7 3.2 1.3 1.2 0.17
students 0.8 1.5 0.8 3.5 1.5 1.3 0.50
Table 3: The average use of identified emotions for all sessions.

4.2 Emotion communication interactions

Between the two tutoring sessions, 167 instances of ECI were identified in the data. Of these instances, 112 (67%) were responded to by the tutor or student with emotional content. Finally, 69 (41%) ECI instances displayed evidence of emotional convergence where participant dyads began using similar emotional styles, content or forms.

Overall, ECIs grew between sessions and the average number of turns for each ECI also increased. However, there are exceptions. Dyad 3 had fewer ECIs (14 to 11) in the second session, but each ECI averaged more turns (2.7 to 4.2). Dyad 6 also had a similar result, except the number of ECIs increased over sessions, while the number of turns decreased. The other four dyads followed a similar pattern of both increasing ECIs across sessions and increasing the average number of turns of each ECI.

Some convergences showed the emergence of extra-emotional content such as in the excitement example in table 4. The teacher begins with one “!” and the student replies with “!!”. By the end of the interaction, the student gives the final emotional celebration by replying in all caps, two exclamation points and the smile emoticon “^^”. A similar result also occurred during emotionally uncertain instances where the teacher was attempting to guide the student towards a language correction. After a few turns of uncertainty and hesitancy, the teacher praises the student with encouragement and “!”.The most common form of convergence was in the use of exclamation marks (!). 5 of the 6 dyads showed convergence, followed by using encouraging content. 4 of 6 dyads began replying to encouragement with encouraging words. Other EC types were dyad specific or shared only between two or three dyads. Hesitancy was the predominant way dyads 1 and 6 expressed convergence.

Encouragement Excitement (!) Hesitancy
  Dyad 1


since they fought on the island, it divided into two pieces


nailed it

They fell into the ocean and kind red monster saved them


cant be better than this


haha wow

full of confidence now!

   Dyad 2


Ok, good!

Let’s look at picture B now.

Can you describe that picture?


Yes!! The man who ride a  motobike hover in front of the car, so I can guess that bike drive horribly.


Great! Yes, it looks like he is a terrible driver!



Dyad 5


good catch… when I type on the computer I misspell too

anything else?


actually, I can’t get anymore..


What about the word motocycle?


Oh, is it motorcycle?



Table 4: Different examples of emotional categories related to teacher/student interaction

4.3 The ecological SCMC environment and action

Using the categories of emotion in Derks et al (2007) and the affordances listed in Lamy and Hampel (2007), it becomes clear that there is a very specific semiotic budget (van Lier, 2000) of the teacher/student ICC in this study’s SCMC environment, listed in table 6. The impact of the synchronous nature of chat impacted how participants explicitly presented their digital physical emotions through emoticons and punctuation.

Dyad 6

C,F,B, D, E, A


Exactly! You’re a super sleuth.

That’s Sherlock.


Next task is…

We will go through the story you told me. You will decide if there is an error in the line, and correct the error.


the man driving his motercycle hit the building and some people came to see if he was okay.


the man driving his mortorcycle hit the car and the car driver and his friend ran to see if he is Okay.

if he was okay.

Table 5: Social expectations and IC misunderstanding related to time.

The pressure to receive a response to EC caused some participants to be impatient. Table 5 shows how the use of an idiom to praise the student, instead of encouraging the student, made it difficult to respond to. In turn, the teacher felt unsure about how to continue without getting a reply to their idiomatic use, as evidenced by their hesitation, “Anyway…”

Google chat affords users a huge number of possible emoticon options and it would be, in principle, possible to very specifically convey the exact emotion a user wants to express. However, in the data, participants generally favored one specific kind of “laugh” or “smile” emoticon (haha and 🙂 ). The social pressures related to the native-teacher/Korean-student authority and unfamiliarity limited the responses of some participants. The written language was similarly affected by the chat environment. It is possible that the writing style, including informality and misspellings are comparably less formal than ACMC teacher/student interactions, such as email.

Physical Written language Social closeness / distance
Time •huge range of possible choices

•No time to keep looking for a “perfect” non-verbal cue.



Similarities to f2f language classrooms?

1-on-1 tutoring?

•Pressures participants to respond more quickly, similar to f2f
Mode •Lack of visual cues increases use of “!”

•Less use of “.”

•Turn-taking emotion



•Non-formal standards

•Visual anonymity,

no increase in negativity

•Teacher/student environment

Table 6: Time and mode affordances in SCMC

This study found that negativity was highly limited between participants. The anonymity of the participants was certainly that of strangers without f2f physical cues, but socially, the participants had important information that allowed them to situate themselves. Before the tutoring sessions, both the teachers and students were made aware that they were going to be part of a tutoring project and the students in this study were aware that their teachers were native-English speakers. This information was used by participants to act in certain roles not unlike traditional teacher-student roles.


5.1 Agentive emotion communication

How the participants were able to use the digital environment to express traditional physical/emotional cues shows how important we view them as tools. An important insight into the CMC teaching environment and expression of emotion however, is the active, willful decision to use emotion.

Compared with f2f interactions where physical emotional cues run parallel with the linguistic content (i.e. smiling while talking, nodding while listening), CMC environments require that the participant to 1) be knowledgeable about how to use “physical” cues in digital spaces and 2) exert their subjective agency to use emotion as communication. In terms of ICC then, part of successful emotion communication is in the use and understanding of shared emotional knowledge. From a CAT perspective, the default linguistic stance is one that optimizes the ability of both interactants to communicate (Hamers and Blanc, 2000). With this in mind, while the environment affords the participants the choice to use English EC such as “lmao”, it never appeared in the tutoring sessions for this study, whereas “haha” was dominantly used by both NEs and KEs.

The desire to build relationships with their partner also led some of the English-speaking teachers to use Korean emotional content, such as “ㅋㅋㅋ”. In Byram’s (1997) terms, this signals 1) intercultural knowledge 2) ICC interaction skill and 3) positive attitudes towards Korean culture. Additionally, the power-relationship not only between teacher-student but so-called native English speakers and L2 English speakers is one of social dominance. The NE’s use of Korean EC styles signals something much different than a KE’s use of Korean EC. Table 7 shows how the NE uses Korean EC to express closeness with their Korean student, who is having difficulty with the English keyboard. Whereas the KE’s use of Korean EC in dyad 1 is possibly an act of social distance, let out due to the frustration of corrective feedback.

KE use of KEC NE use of KEC
Dyad 1


“and kind red monster saved them”?


the kind ~


great ^^



Dyad 5


Oh, I did misspelling a lot. I am not used to english keyboard. I have to change uncomfortable and motorcycle.


Don’t worry, I am used to an English keyboard and I always make mistakes ㅋㅋㅋ

Table 7: use of Korean EC conventions in SCMC

5.2 Closeness and emotional regulation

In attempting to determine whether or not teacher/student pairs exhibit social closeness or distance under CAT (Giles, 2008), it became difficult to know whether or not instances of EC were actually emotionally distant or close. The lack of a comparable post-tutoring survey limits the ability of this study to determine whether or not there is a correlation between specific types of emotional content and satisfaction with tutoring interactions.

However, using Matsumoto et al’s (2007) concept of emotional regulation, we can evaluate the EC interactions in this study as attempts to maintain a relationship, even if the interactions are stressful or uncertain. While it might seem that examples such as dyad 6 in table 5 suggest that the pair were becoming more emotionally distant, later on in the session the student expresses satisfaction with the tutoring session overall saying, “I enjoyed this task”. Hesitant reactions and use of “…” or uncertainty are not necessarily markers of emotional distance. Instead, they might suggest emotional vulnerability and signal to the other participant that they should signal closeness or also signal uncertainty. Hesitancy, it seems, can be a signal of trust and respect for the perceived social roles involved.

5.3 The role of the teacher

It is worth noting that no teacher/student pair exhibited any of the behavior of negativity found in other CMC studies (Mckee, 2002) that generally focus on peer ICC interactions. The social roles already mentioned played an important part in this, but not all dyads behaved in similar ways. Across the six dyads, three interactant types seemed to emerge. The most emotionally similar dyads seem to interact in more egalitarian ways, or as if they were friends. Three of the dyads seem to exhibit guide/follower roles, where the teacher uses encouragement and hesitancy to guide the student through learning, but they aren’t seen as traditional teachers. Finally, one dyad exhibited a much more traditional teacher/student relationship, where the student showed a lot of uncertainty, hesitancy and deferred to the teacher.

Importantly, across all interactant types, participants seemed to demonstrate 1) adequate emotional regulation and 2) Inter-cultural and SCMC knowledge. It would be reasonable that perhaps the teacher in the traditional teacher/student relationship might feel uncomfortable due to the lack of positive emotional cues. However, the teacher seems to accommodate the student’s hesitancy and continues to regulate their emotional presentation and use encouragement to teach.

As a teacher, the results of this study suggest that specific kinds of EC from students should not lead the teacher to assume social distance or that perceived social distance is harmful. Teachers interact with many more students than students will with teachers, so it is important that they be emotionally competent and regulate well. What may seem like an emotionally distant student may in fact just be different social expectations and pressures that they feel. It is perhaps a popular feeling that teachers are already emotionally well-developed and know how to interact in ICC situations, but in fact teachers can be just as ignorant as anybody else, and they should prepare themselves with the same toolkit as students, in order to successfully navigate their interactions.


This study attempted to investigate the kinds of emotion communication found in teacher/student dyads in intercultural SCMC environments. It found that SCMC does not greatly change the type of emotional content in teacher/student interactions, but that style and writing conventions are reappropriated and used for new uses in order to fill the gap in physical and other nonverbal cues. The social expectations related to authority and nativeness, while challenged in many areas, need to be taken into consideration when teachers attempt to communicate emotionally with students and accommodate the various expectations that students have.

6.1 Limitations and future research

Not present in this study was a post-tutoring survey that could give the personal feelings of the participants of emotional closeness or distance. This is an important item which would help determine whether or not the interactions really were evidence of accommodation or not. Additionally, while the focus of this study was specific to the intercultural aspect of non-Korean teachers and Korean students’ comparison with Korean teachers and Korean students might shed light on specific interactant cultural differences that are not entirely clear in this study alone.

Additionally, this study examined just two, one-hour, tutoring sessions. The development of the teacher/student relationship over a longer period of time would shed more light how and how much intercultural teacher/student pairs converge or diverge in their use of language and emotion communication.


Blommaert, J. (2010). The sociolinguistics of globalization. Cambridge: Cambridge University Press.

Bretag, T. (2006). Developing ‘third space’ interculturality using computer-mediated communication.Journal Of Computer-Mediated Communication, 11(4), 981-1011.

Byram, M. (1997). Teaching and assessing intercultural communicative competence. Clevedon: Multilingual Matters.

Creswell, J. (2003). Research design. Thousand Oaks: Sage Publications.

Derks, D., Fischer, A., & Bos, A. (2008). The role of emotion in computer-mediated communication: A review. Computers In Human Behavior, 24(3), 766-785.

Gibson, J. (1977). The theory of affordances. In J. Gieseking, W. Mangold, C. Katz, S. Low & S. Saegert (2014), The People, Place, and Space Reader (1st ed., pp. 56-61). New York: Routledge.

Giles, H. (2008). Communication accommodation theory. In L. Baxter & D. Braithewaite, Engaging theories in interpersonal communication: Multiple perspectives (1st ed., pp. 161-173). Thousand Oaks: Sage Publications.

Hamers, J., & Blanc, M. (2000). Bilinguality and bilingualism. Cambridge: Cambridge University Press.

Lamy, M. N., & Hampel, R. (2007). Online communication in language learning and teaching. Basingstoke: Palgrave Macmillan.

Matsumoto, D., Leroux, J., & Yoo, S. H. (2005). 学術講演会 Emotion and intercultural communication. 西院大会学部紀要, 99, 15-38.

McKee, H. (2002). “YOUR VIEWS SHOWED TRUE IGNORANCE!!!”: (Mis)Communication in an online interracial discussion forum. Computers And Composition, 19(4), 411-434.

Meskill, C. and Anthony, N. (2010). Teaching languages online (2nd ed.). Retrieved from

Orengo Castellá, V., Zornoza Abad, A., Prieto Alonso, F., & Peiró Silla, J. (2000). The influence of familiarity among group members, group atmosphere and assertiveness on uninhibited behavior through three different communication media. Computers In Human Behavior, 16(2), 141-159.

Pasfield-Neofitou, S. (2013). “Digital natives” and “native speakers”. In F. Sharifian & M. Jamarani, Language and Intercultural Communication in the New Era (1st ed., pp. 138-160). New York: Routledge.

Scissors, L. E., Gill, A. J., Geraghty, K., & Gergle, D. (2009). In CMC we trust: The role of similarity. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 527-536). ACM.

Tucker, M. L., Gullekson, N. L., & Esmond-Kiger, C. (2014). Accounting for EI: Does emotional intelligence predict greater intercultural growth?. Journal of International Business and Cultural Studies, 8, 1-1.

van Lier, L. (2000). From input to affordance: Social-interactive learning from an ecological perspective. In J. Lantolf, Sociocultural Theory and Second Language Learning (1st ed., pp. 245-261). Oxford: Oxford University Press.

Appendix – ICC Journal

  1. Find each instance of emotion communication in the transcript. Instances include the initiating emotion and the responding emotion.
  2. For each instance, complete the chart below.

Which participants are involved?

Who initiated emotion communication?

Which emotion?

Did their partner respond emotionally? How? Turns? Was there evidence of social closeness or distance?


  1. Of the total emotion communication instances, how many prompted responsiveness?
  2. How many shared emotional content?
  3. Linguistic development. First, look at the chart above. Did the same emotion communication occur more than once? In the first instance, what linguistic cues were used and what was the response. How many times did the same type of emotion communication happen?
dyad Emotion communication shared Number of instances


  1. Changes across tasks?
dyad Emotion communication shared Change in number of instances?


  1. For questions 5 and 6 above, how many instances of linguistic closeness/distance are there?
  2. Look at the time and mode affordances
    1. How is the affordance of time playing positive or negative roles in emotion communication?
    2. How is the affordance of mode playing positive or negative roles in emotion communication?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.