Turn on Your Camera, Foo : Slang and Visual Cues in the Classroom

Jiajun Weng, Chris Lam, Christine Chang, Terri See Lok Ho, Wei Lin

Have you ever wondered whether understanding what your classmates are saying and the seeing their cameras is essential to succeed in the course?

You’re not alone.

During this special period, education has primarily moved on to online. Many international students from UCLA taking online courses claim that they feel alienated in the class because they cannot see their classmates when their classmates are talking, and they sometimes cannot understand the online slang used by their classmates. Does the usage of online slang and lack of visual cues truly impact their learning experience?

For finding out the answer to this question, we conducted a study to investigate how the use of slang and the lack of visual cues contribute to international students’ comprehension difficulties and their feelings of alienation. The survey sample comprised entirely of UCLA students. By analyzing the data, we found that interestingly, their feeling of alienation was not affected by usage of online slang nor lack of visual cues. Furthermore, we found that their comprehension was not associated with inclusiveness. That is, it shows that one can still succeed in the class even if one feels alienated.

Introduction

International students in an English-speaking country such as the United States face various challenges related to language. For instance, they struggle with the use of slang and cultural references in a classroom setting. In Bradford’s research, he found that “Teaching colloquial speech in any language can be important for acquisition and assimilation into the language’s cultural group” (Bradford, 2010). In a separate study, Albalawi found that some L2 learners indicate learning slang is helpful for students to fit in socially in college and gain confidence (Albalawi, 2014). Both articles demonstrate that learning the slang of other cultural groups is a crucial tool for L2 learners to master if they want to become more assimilated. However, some of these difficulties in comprehension can be overcome by implicit cues such as facial expressions and gestures (Sueyoshi & Hardison, 2005).

A comment about Figure 1 below showing the breakdown of UCLA students and their experience with English. Moving onto online learning platforms during the global COVID-19 pandemic, international students face new challenges like adapting to the lack of visual cues — such as facial expression and gestures — as well as the colloquial way people speak during Zoom lectures. As a consequence, international students’ ability to comprehend course materials may be compromised by the lack of social cues. With limited understanding of course material, these students could subsequently feel disconnected, or even isolated from the class, and hence disengaged with the course.

International students who engage primarily on non-English social media platforms, such as Wechat, Weibo or KakaoTalk, may have found it more difficult to navigate higher education in this virtual environment. Within this experimental study, we investigated how the use of slang and the lack of visual cues contribute to international students’ comprehension difficulties and their feelings of alienation. Specifically, we expected to find increased feelings of alienation and reduced engagement among international students in the face of online English jargon and little visual cues. However, we hypothesized that the use of slang should not significantly impact students when visual cues are present in the recorded lecture because non-verbal communication can be an important source of motivation and concentration for students’ learning as well as a tool for taking and maintaining attention (Zeki ,2009).

Figure 1: UCLA students’ distribution

 

Collecting Data: Setting up a Classroom

An experimental study was conducted to test our hypotheses about how the manner of people’s speech during the lecture and visual cues (i.e., facial cues and gestures) interacted with each other to influence international students’ understanding of the course materials and their feeling as a member of the class. In the current experiment, we showed our 16 international student participants one of the four Zoom lecture recordings in which we systematically varied the manner of speech of people in that class, as well as the presence of visual cues. To manipulate people’s manner of speech during class, the student confederates discussed the class material in standard English or in a colloquial manner that involved the use of English slangs, like “btw” or “hella”. To manipulate the presence (or absence) of visual cues such as facial cues and gestures, confederates in the current class video will either turn on or off their camera and showed their face and hand movement during the lecture recordings. Please see Table 1 for a demonstration.

Table 1: Matrix of variables and samples of corresponding experimental script

 

After the participant watched one of the four mock zoom lecture recordings, they were instructed to complete a questionnaire that assessed their understanding of the lecture content, which is about this basic psychological phenomenon called the cognitive dissonance theory. Besides the objective measure of participants’ understanding of the class material, their subjective perception of how well they understood the lecture was also assessed. Finally, we measure how much these participants feel like a member of the class and the likelihood of engaging with the lecture if they were present in the Zoom meeting room.

Results and What They Mean

Figure 2: Video On and using Slang trail; participants’ feeling of alienation

 

Our findings supported the initial hypothesis that having video on in these online lectures affected students’ level of comprehension. However, there wasn’t a statistically meaningful difference in feelings of exclusion. In particular, the analysis showed that there was a meaningful difference between the results of the survey question regarding subjective comprehension conducted with the students who watched the lecture with video and without video, regardless of whether there was slang or standard English used. However, even by looking at Figure 2 and Figure 3, it is clear that students felt excluded either way.

Figure 3: ’Video Off and using Slang trial; participants’ feeling of alienation

 

The same was true regarding our initial hypotheses about slang usage. The study showed a statistically meaningful difference in level of self-reported comprehension, but not on the feelings of exclusion. Visually comparing Figures 2 and 4 shows that the responses to the question about alienation were not meaningfully different when video was on vs. off in Figures 3 and 5. Even a cursory review of the results of all the survey questions that attempted to measure feelings of exclusion and alienation showed high levels across the board, boding negatively for online classes as a whole.

Figure 4: Video On and using Standard English trial; participants’ feeling of alienation

 

Figure 5: Video Off and using Standard English trial; participants’ feeling of alienation

 

And perhaps the most important result came from comparing the interaction factors of video and slang in the comprehension question. What our study found was that while there is a statistically meaningful difference in individual comprehension in response to both factors, i.e., video and slang, there was not significant interaction between them when it came to self-reported levels of comprehension. That is to say, contrary to our initial hypothesis, a factor like having video on doesn’t necessarily interact meaningfully with the differences caused by slang usage in comprehension.

Figure 6: The rate of participants correctly answering the quiz questions

 

However, when it comes to the actual analysis of the answers to the quiz questions, not just self-reported comprehension, there is a noticeable interaction factor. Figure 6 shows that when slang is used, the presence of video had a significant impact on actual comprehension as measured through the quizzes, whereas video had less impact when standard English was used. 

There are, of course, various factors that could be complicating this kind of analysis. The subjects chosen for the mock lesson, were it more or less visual, may have more of an effect on how these two variables interact. The length of the lesson may have an impact on all these variables depending on how often it becomes relevant that video is used or not. This study is not necessarily definitive but poses some important questions on how all of these variables can be utilized by educators in aiding comprehension and limiting alienation in classrooms.

Conclusion: The Classroom and Beyond

All in all, more research should be done with regards to the virtual learning environments that most of the world was thrown into due to the pandemic. There may be many key improvements to education in general, whether online classes are here to stay for a while or not. From our initial hypothesis that English slang negatively impacts international students’ engagement with and understanding of classroom material, we find that comprehension may be hindered by slang usage and a lack of visual cues, independently; however, international students seem to feel like they do not “fit in” with the class regardless of these variables, and their aptitude does not seem to suffer because of that in general.

We live in an ever growing technologically dependent society, yet online meetings can often feel like an obstacle and/or a divider when compared to in-person classes. In the article “Depression and Everyday Social Activity, Belonging, and Well-Being,” Michael and Todd stated that “When people experience positive social interactions, they should be more likely to feel a sense of belonging.” Alongside virtual meetings in the workplace, there is a lot that can be done by the hosts to improve distracting and frustrating video calls. Based on this small study, we recommend professors and teachers to encourage students to turn on their videos, with the caveat that there may be personal and privacy challenges. We can say that there may be evidence that doing so will help students’ comprehension of the material. We also suggest addressing slang and jargon when it arises in the classroom, making sure to at least clarify rather than exacerbate what may negatively impact some students’ learning outcome.

There are an endless number of questions to be asked in the realm of education research, with regards to both online and in-person mediums. Perhaps this experiment may be repeated with a live virtual classroom setting to really capture engagement and chat-box interaction data. Furthermore, there is something to be examined in asynchronous learning, i.e., these pre-recorded lectures in the study that subjects independently and asynchronously watched. In a pandemic that generates so many struggles, personally and in education, there is the possibility that Zoom lectures are a breakthrough to education access the world needs; we just need to optimize and adapt to it, rather than conceding at its shortcomings.

 

Further info

The PowerPoint form of this blog entry

A TedTalk which talks about the relationship between inclusiveness and your manner of speech

 

 

References

Albalawi, A.S. (2014). Saudi L2 learners’ knowledge and perceptions of academic English slang. [Order No. 1566835]. Southern Illinois University at Carbondale.

Bradford P.B. (2010). The acquisition of colloquial speech and slang in second language learners of English in El Paso, Texas . [Order No. 1484150]. The University of Texas at El Paso.

Steger, M. F., & Kashdan, T. B. (2009). Depression and Everyday Social Activity, Belonging, and Well-Being. Journal of counseling psychology, 56(2), 289–300. https://doi.org/10.1037/a0015416

Sueyoshi, A. & Hardison, D.M. (2005). The Role of Gestures and Facial Cues in Second Language Listening Comprehension. Language Learning, 55: 661-699. https://doi.org/10.1111/j.0023-8333.2005.00320.x

Zeki, C. P. (2009). The importance of non-verbal communication in classroom management. Procedia-Social and Behavioral Sciences, 1(1), 1443-1449.

Read more

Code-Switching Between Mandarin Chinese and English: Do You Use “lol” or “xswl”?

Wenqian Guo, Sum Yi Li, Yichen Lyu, Sok Kwan Wong, Yingge Zhou

Code-switching has become increasingly common as globalization allows international exchanges across cultures to take place more frequently. And as studying abroad becomes more accessible to students around the world, more speech communities with distinctive code-switching patterns are being formed. As we pondered the topic for our research project, we looked around and realized that not only are the majority of our group members native Mandarin speakers studying in the US, but collectively we also belong to this wider speech community that tends to code-switch between Mandarin and English. We could not help but wonder — do local students in China talk like us at all? And is there a reasoning behind the way we talk? It is these questions that formed the basis of our research.

For the project, we narrowed down our research to focus on just Internet slang used on WeChat, China’s answer to WhatsApp. Through our proprietary survey and by combing through chat history we collected from our participants, we discovered some very interesting findings. Continue reading to find out how and why Mandarin-speaking international students in the US code switch on WeChat.

Introduction and background

Chinese students who study in the U.S. often code switch between Mandarin and English and our project was aimed at examining the motivations behind the phenomenon.

We specifically looked into our subjects’ texting patterns on messaging app WeChat and compared them to Chinese students either residing in China or other non-English speaking countries in an attempt to confirm our assumption that code-switching is prevalent among Chinese students in the US. We also surveyed the students and asked for reasons behind their choice of words. We hypothesized that convenience as well as a desire to appear foreign-educated are what motivated the code-switching.

We based our project on Luke’s (1984) study on language mixing in Hong Kong, where English words are often inserted into Cantonese conversations. Luke (1984) concluded that code-switching in Hong Kong is partly pragmatically motivated (when the objects being discussed do not have Chinese translation) and partly socially motivated (when the individuals want to identify as better educated and westernized).

We extended Luke’s (1984) study to cover Mandarin Chinese, which is spoken in Mainland China. Presumably English mixing is more prevalent in Hong Kong because it is a former British colony. Code switching is not common among locals in China, but it is observed among individuals who have exposure in English-speaking countries.

Methods

The target population for our project consists of two major groups: college students who study in the US and college students who study in their home country China. Both groups of students are native Mandarin speakers. Our data collection was divided into two parts. First, we created a survey to ask both groups of students to select from provided word choices under different text conversation scenarios and provide us with a reason behind each choice. The participants were also asked to specify how frequently they would code-switch in their daily conversations with friends and family on WeChat. Second, we collected a series of chat history based on three main topics, namely schoolwork, casual conversations and sensitive topics. For the survey, we interviewed a total of 17 students from UCLA as our sampling for the groups of students who are foreign educated. We also interviewed a total of 6 college students who are based in China.

Results

For the purpose of discussion, our project referred to college students who study overseas as “US students” and those who study in China as “local students”.

School Work Survey Question 1: Which word would you use when you want to talk about a homework assignment that will be due soon. For example, “我明天有个作业___”. (Tomorrow I have some homework__)

For the first survey question regarding homework assignment, all of the US students chose the English word “due”, while the majority of the local students chose the equivalent Chinese words “要交”. Even though most students from both groups attributed their choice to a similar reason, which is language habit, from the perspective of the US students, “language habit” refers to a way to try to assimilate into the American culture, while in the context of the local students, it is more of an innate and natural habit.

School Work Survey Question 2: Which word would you use when you want to unenroll a class you have registered before? For example, “这节课不符合我的时间表, 我想__这节课”. (This class doesn’t match my schedule, I want to __ this class.)

For the second survey question relating to unenrolling classes, all of the US students chose the English word “drop”, while the majority of the local students picked the Chinese equivalent “退选”. However, both groups have different reasons behind their word choice. The majority of the US students said they chose the word “drop” because there is no equivalent translation in Chinese. On the other hand, the local students preferred using Chinese due to language habits influenced by their friends and family.

Based on the reasons the participants provided, it appears that being in different environments and different speech communities are the main reason students develop different language habits. Besides, the source of learning also influences their word choices. The US students tend to find it hard to find Chinese translation for words related to schoolwork since they learned these words in an English-speaking environment.

Casual Conversation Survey Question 1:  Which word would you use if you want to express laughter or something that is funny when you are chatting with your friends.

When asked how they would express laughter when texting, the majority of the US students chose the Chinese words “哈哈哈哈” (“Hahahaha”), while the majority of the local students picked another Chinese phrase “笑死我了” (“I laugh to death”). Even though both groups of students used Chinese to express laughter, each side has their own reason for the specific choice. The US students said they preferred “哈哈哈哈” (“Hahahaha”) as a language habit, while the local students preferred “笑死我了” (“I laugh to death”) because they believed the expression could better demonstrate their emotions and the situation. The results showed that Mandarin-speaking college students preferred to use Chinese when expressing laughter regardless of where they are studying.

Casual Conversation Survey Question 2: How often do you often replace words in a sentence from Chinese to English when you are chatting with your friends? For example, “我一会儿有个meeting or presentation”, “让我来表演一段rap”.

When asked how frequently they code-switch between Chinese and English when texting their friends and family, the majority of the US students said at least once every one to two days. Meanwhile, half of the local students said they seldom code-switch — only at least once every few weeks or months during their daily conversations.

Sensitive Topic Survey Question 1: What kind of curse words would you use most frequently when you are chatting with your Mandarin speaking friends?

When asked which curse words they most frequently use, both groups of students chose “卧槽/我操/我靠” (roughly translated as “Damn it/Fuck”) but for different reasons. The US students said this option best describes their feelings, while the local students said it sounds less harsh than the other choices.

Sensitive Topic Survey Question 2: Which word would you use when you need to discuss something that is related to the issue of sexual assault. For example “你上个月有听说那条新闻吗?有个女生被__了”. (Did you hear the news? A girl was __)

When asked what words they would use to say “rape”, the majority of both the US and local students chose the Chinese words “强奸”, instead of “rape” in English or the abbreviation “QJ”. Most local students said they did not worry whether the word sounds too direct or inappropriate but would rather want to just say what really happened. Similarly, the US students said they felt more comfortable with the Chinese words.

Our survey: English version and Chinese version

Chat History Analysis

The chat history was obtained from Chinese students studying in the US. The words marked red were originally in English, while those in black are translations from Mandarin.

This is a conversation between two participants about the recent US election. P1 said they expected people to stop protesting in two days. Instead of using the Chinese words (“游行”) for protest, P1 code-switched from Mandarin to English. Since protests are rare in China, and the Chinese words for protests are rarely used, presumably it is easier for the participant to just use the English word when texting.

Marked code-switching is observed when P1 expresses a slight disagreement with P2. As P2 feels empathetic to Trump, P1 emphasizes with English to express that they don’t feel sad for Trump’s loss.

Another interesting point is that P2 referred to Trump as “Grandpa Trutru”, his Chinese nickname. Chinese people sometimes use nicknames to refer to important politicians, which is likely stemmed from China’s censorship on sensitive topics. Discussions about certain politicians are considered highly sensitive, so to avoid censorship, they come up with nicknames for the politicians. For example, Trump is also known as “懂王” (“the king who knows it all”), while Biden is “睡王” (“the sleepy king”). The use of nicknames exudes a sense of humor as well as dials down the seriousness of the discussion of political issues.

In casual conversations, code-switching again lends convenience and gives emphasis. It also constructs a common identity among people in the conversation. In the above conversation, the participant is telling a story about their roommate being forcefully taken away to the hospital after answering routine behavioral questions wrong, which is a well-known cultural shock among Chinese students studying in the US. As this routine is not performed in China, the participant constantly code-switched on noun and verb phrases for convenience. Also, all participants of this conversation are Chinese international students. The code-switching builds a common identity among them because the participant expects everyone to know the consequences of answering yes to routine behavioral questions. The last two sentences are examples of marked code-switching that emphasize on the participant’s disbelief: The participant is surprised that his roommate answered yes. It is a final revelation of the ending to this anecdote.

Similarly, in the above conversation, code-switching again serves as building a common identity among Chinese students in the US. The participants constantly chose to use short English vernaculars, such as “yes,” “go,” and “yea.” This indicates that they have been immersed in an English-speaking environment, so when expressing agreement and excitement, they tend to code-switch to English. Also, P1 ignored all English spacing in the conversation. This is because when typing English with the Chinese keyboard, adding spacing can be time-consuming. For convenience, P1 simply ignored the spacing during texting. Even so, P1 chose to respond in English, demonstrating that it is more natural for them to use these English vernacular phrases.

In academic scenarios, code-switching almost entirely occurs in jargon. By using English jargons such as “peer interaction” and “mutual engagement”, the participants demonstrated their educational background as foreign.

Discussion and Conclusion

From the aforementioned data findings, it can be concluded that Chinese students who study abroad and those who only study in China show different patterns in their language usage. When it comes to schoolwork and casual conversations, insertions of English words into Chinese sentences and code-switching between English and Chinese are mainly observed among students who study abroad, while less so is seen among the domestic students. This difference can likely be attributed to a lack of translation equivalence, as many school work-related words are only applicable in the US. The same goes for casual conversations. It is therefore not difficult to understand why the US students would tend to code-switch to English when there seems to be a lack of translation equivalence in Chinese.

Besides, the needs for social and emotional expressions evoked by surrounding cultural environments and contexts also contribute to the different language patterns and code-switching. In China, a conservative country both culturally and politically, pro-government language is encouraged, while the freedom of speech is repressed. This repression has in turn cultivated stronger needs for expressing emotions among the local students, leading to their choice of more confrontational and direct wordings when discussing sensitive topics. On the other hand, a higher frequency of code-switching among the US students revealed their needs to be identified as foreign-educated and share common identities with other speech participants of similar backgrounds.

Our findings point to the important roles that social surroundings and the kind of language encouraged within these environments could play in one’s speech patterns and code-switching. That being said, the language choices that the participants made are not entirely dependent on their own characteristics but are rather choices commonly negotiated by one’s surrounding social context as a whole. This contextual-based understanding therefore sets a reminder for future conversation analysis and sociolinguistic study, that one’s speech patterns could not be analyzed and identified without incorporating the nature of the surrounding speech contexts and cultural environments.

 

More on the topic:

Video about attitudes toward code switching in China

Paper on Chinese-English code-switching in conversations

References

K.K. Luke (1984), Expedient and Orientational Language Mixing in Hong Kong, York Papers in Linguistics 11, 191-201

Read more