A New Online Co-Viewing Experience:
The Effect of Danmaku On Dutch And Chinese Young Adult Audiences Author: Wanlu Fu
Student ID-card: 12027391 Master’s Thesis
Graduate School of Communication Master’s Program Communication Science Entertainment Communication
Supervisor: Susanne E. Baumgartner Word Count: 6509
Abstract
Danmaku is a new form of comments on online video sites with the co-viewing feature. Danmaku appears as overlaying the video screen and is synchronized to the playback time of the video. Despite of the success in China, Danmaku is rarely be known to Dutch people. Therefore, the aims of this study were to investigate (a) the effects of Danmaku (i.e., perceived social presence, entertained feelings, perceived information value of the video, feelings of irritation) on young adult audiences, and (b) how do these effects differ between Dutch and Chinese young adult audiences. 189 (aged 18–30) respondents were recruited from personal connections (N = 131) and from an online research company website called the Prolific (N = 58). A survey-embedded online experiment was conducted via Qualtrics. The findings show that Danmaku can provide a higher level of perceived social presence and feelings of irritation to young adult audiences. Moreover, Chinese young adult audiences experienced higher levels of social presence while Dutch young adult audiences felt more irritated during Danmaku viewing. However, the effects of entertained feelings or perceived information value of the video were not confirmed by the current study. More research is needed on the effects of Danmaku in the future.
Keywords: Danmaku, social presence, information value, entertainment, Dutch and Chinese young/emerging adult
A New Online Co-Viewing Experience:
The Effect of Danmaku On Dutch And Chinese Young Adult Audiences Introduction
Online video viewing has become increasingly important in modern daily life. Online video viewing has increased by 34% worldwide by the year of 2017 (Martin, 2018). Taking into consideration the growing needs of co-viewing and social interaction in online video viewing experience (Haridakis & Hanson, 2009), young adults became more engaged in online video viewing and in sharing the content with others (Bondad-Brown, Rice, & Pearce, 2012).
‘Danmaku’ (Yan, Cha, Cho, & Hwang, 2019), a new online video commentary feature providing co-viewing experience, which is also named as Bullet Screen (Djamasbi, Hall-Phillips, Liu, Li, & Bian, 2016), has thrived in Japan and China. Compared to forum comments (e.g., forum comments on YouTube.com, see Appendix A), synchronicity and anonymity are the two main specific characteristics of the Danmaku (Wu, Sang, Zhang, & Huang, 2018). Synchronicity means that Danmaku enables the comments to appear
overlaying the video screen and synchronized with the video streaming (See Appendix A). Viewers can send or read the comments that are evoked by the video content of the current time point during the viewing session (Chen, Gao, & Rau, 2017). Anonymity means that user-account names would not be presented with the Danmaku posting, only showing the content of comments on the screen (Wu et.al, 2018).
In 2006, the first Danmaku video site appeared (i.e., nicovideo.jp) in Japan. Later on, Danmaku has become indispensable for many young Chinese viewers to watch online videos (Chen et.al, 2017). After an incredible popularity of Bilibili.com (i.e., Chinese Danmaku video site) among young adult users, almost every Chinese video website (e.g., Youku.com, IQiyi.com, etc.) has adopted the Danmaku feature (Chen, Gao, & Rau, 2015). Moreover,
other Chinese media have also tried to combine the content with the Danmaku feature (e.g., Danmaku advertisement, Danmaku film, etc.). The viewing experience of Danmaku video has already obtained a big success in China, but Danmaku has not been introduced to the Dutch market yet. Thus, I believe that there are potential possibilities for emerging and booming in the Netherlands.
Although Danmaku has already been very popular in China, the effects of Danmaku are not well studied yet. Moreover, the effects of Danmaku on young adult viewers that are unfamiliar with Danmaku are not clear either, taking into consideration the cross-culture perspective. Besides the studies on media capabilities and features of Danmaku (Liu, Suh, & Wagner, 2016), previous studies have mainly explored and concluded the motivations and reasons for watching the Danmaku (e.g., Chen et.al, 2015; Chen et.al, 2017) among Chinese young adult viewers. Entertainment needs, seeking information and feeling of being
accompanied are the three main reasons for young people to enjoy the Danmaku (e.g., Chen et.al, 2015; Chen et.al, 2017). In the meantime, studies also show that Danmaku has some negative characteristics (e.g. information abundance) that influence the popularity of Danmaku (Chen et.al, 2015).
Inspired by the results of the previous studies of the Danmaku, the present study aims to investigate if Danmaku would have indiscriminate effects (i.e., perceived social presence, entertained feelings, perceived information value of the video, feelings of irritation) on viewers, and to provide a first comprehensive overview of the differences between Dutch and Chinese young adult users.
Thus, the main research question is:
RQ. How does exposure to a Danmaku video affect the viewer’s viewing experience (i.e., perceived social presence, entertained feelings, perceived information value of the
video, feelings of irritation), and how does this differ between Dutch and Chinese young adult audiences?
Theoretical Background Young adult and Danmaku video
Young adults or emerging adults have been defined as people aged between 18 to late 20s (e.g., Arnett, 2004; Coyne, Padilla-Walker, & Howard, 2013). Although Arnett (2004) defined that emerging adulthood is between the age of 18 to mid 20s, there is no specific common definition for emerging adult in academia. Therefore, young adults aged between 18 and late 20s will be studied in this research.
Young adults today grew up in a saturation of media, and their leisure time is mostly occupied with media use (Coyne et.al, 2013). Young adult internet users (i.e., 18-30 years old) continue to be the heaviest consumers of online video content (Purcell, 2010).
Autonomy, intimacy, and identity are the three key words for young adult media usage (Coyne et.al, 2013).
Danmaku is popular among young adults (Chen et.al, 2015). For instance, 92.8% of Bilibili.Com’s users are under the age of 30 and 53.6% users are between the age of 20 and 24 (Team, 2020). Autonomy, intimacy and identity can be provided by Danmaku to young adults. First, users can independently choose to turn on or turn off the Danmaku. In addition, users can also selectively choose to display the Danmaku according to different needs (e.g., shielding the Danmaku with certain keywords, or only displaying the Danmaku that appears at the top of the screen), and they can even adjust the appearance transparency of the Danmaku (Fang, Chen, Wen, & Prybutok, 2018). Second, Danmaku provides users with an experience of co-viewing, which creates a warm, sociable and intimate online video viewing experience (Chen et.al, 2015). Third, a sense of belonging to a community can be created by
Danmaku (Chen et.al, 2015), which enables the users to gain a certain group identity. Therefore, the current study is mainly focused on the young adult audiences.
Perceived social presence during Danmaku video viewing
Sociability is the key feature of Danmaku (e.g., Chen et.al, 2015; Fang et.al, 2018). Unlike the interaction experiences (i.e., reply and rate to the comments) provided by the traditional forum comments (Siersdorfer, Chelaru, Nejdl, & San Pedro, 2010), Danmaku provides a different way of the online social experience. By being synchronized to the playback time, Danmaku provides viewers with a pseudo feeling of participating in a real-time chat during the video viewing session (Fang et.al, 2018). For instance, when reading or sending comments related to the certain content of the playback time, Danmaku works like ‘social TV’ (i.e., using different devices to share and chat during the TV viewing session, Chorianopoulos & Lekakos, 2008) but with only a single device. Moreover, viewers can also watch the comments sent by previous viewers and this asynchronous experience strengthens the engagement of real time feelings (Zhang, Qian, Pi, & Yang, 2019). Thus, spatially separated viewers can experience the co-viewing environment (Lin, Huang, & Cordie, 2018) and the parallelized experience (Liu et.al, 2016) together by Danmaku.
Danmaku enables viewers to experience ‘warmth’ during online video viewing. The parallelized and co-viewing experience in the online video viewing context indicates the feeling of being able to share the ideas with lots of others during the viewing time. Firstly, “sense of belonging” and “feeling belong to a group” can be experienced during the Danmaku viewing (Chen et.al, 2015). For instance, the density and the content of the Danmaku creates a virtual community for the viewers (Chen et.al, 2015). Secondly, when reading others’ Danmaku, users have an instant and easy way to get others’ ideas, and it strengthens the feeling of being accompanied (Leng, Zhu, Wang, & Gu. 2016).
Social presence. Social presence refers to the degree to which a person is regarded as a "real person" (Aragon & Aragon, 2003) and the degree of perception of being with others in mediated communication of the people (Oh, Bailenson, & Welch, 2018). Within CMC (i.e., computer mediated communication), the feelings of sociability, warmth, and intimacy contribute to the online social presence (Djamasbi et.al, 2016). Due to the features of Danmaku and in line with the previous studies (e.g., Fang et.al, 2018), the concept of social presence was adopted in this study. Since Danmaku have the capability to provide audiences with the sense of social presence, the study hypothesized:
H1. Danmaku video will lead to greater experiences of social presence than a traditional video for young adult audiences.
Entertainment effects of Danmaku during video viewing
Being able to feel entertained is very important for people which is the reason why
they watch TV and online videos (Bondad-Brown et.al, 2012). Danmaku is made to be pleasure-oriented (Fang et.al, 2018), and previous studies show that users generally using Danmaku for fun (e.g., Chen et.al, 2015).
Hedonic content and the content highly related to the current playback time provide for the feeling of being entertained during the viewing. First, viewers can enjoy the hedonic Danmaku. For instance, Chen, Gao, and Rau’s (2015) exploratory study shows that some of the participants said they enjoyed the humorous comments when they were watching Danmaku videos. Even when it comes to the Danmaku with poor quality videos, audiences can still have fun from ridiculing the content with others using Danmaku or just enjoying the Danmaku itself (Chen et.al, 2017). Second, the synchronicity of Danmaku to certain playback time strengthens the feelings of being entertained (Chen et.al, 2017). A feeling of being involved in instant chatting can be perceived by reading the comments related to the current playback time. The current study mainly focuses on the general entertainment effect that
Danmaku can provide, aiming to fill in the gap to see whether Danmaku can have an effect on general users including users who don't use it often.
H2. Danmaku video will lead to greater experiences of feeling entertained than a traditional video for young adult audiences during the video viewing.
Perceived Information value of Danmaku video
Danmaku provides the audiences with additional information (Chen et.al, 2015). Considering the fact that comments appear simultaneously across the screen, comments can create overall perspectives, or many details which are hard to recall later (Chen et.al, 2015). The biggest difference between the normal forum comments of the Danmaku is its feature of synchronicity with playback time point, which enables the audiences to get the information at the same time when they are watching the video instead of searching for it later. For example, viewers can instantly know what the background music of the video is when some other viewers send the name of the music, by simply presenting “BGM: XXX” (Chen et.al, 2015). They can also understand some background knowledge or explanations of the story telling in the video when other people share it within the Danmaku. It is worth mentioning that when audiences are watching a foreign video without translated subtitles, some users even use the Danmaku to present the translated subtitles (Chen et.al, 2015).
Information value theory. Information value theory indicates the level of information involved in communication process (Howard, 1966). During the media
consumption, information is conveyed by the media to the consumers. The utilitarian aspect plays an important role in media engagement (Davis Mersey, Malthouse, & Calder 2010). Previous studies revealed that a lot of audiences who watched the Danmaku can get additional information. Thus, in this study, information value theory is introduced to see if the Danmaku can make audiences feel like they get more information related to the video, so that they can get a deeper understanding of the video content.
H3. Danmaku video will lead to greater experiences of perceiving information value of the video than a traditional video for young adult audiences.
Feelings of irritation during Danmaku video viewing
Although Danmaku has become successful to some extent in China, there are also many Chinese young adult audiences that don't really enjoy the Danmaku because of the negative feelings caused by the Danmaku (Chen et.al, 2017). Information abundance and the aesthetics are the two main negative features of the Danmaku (Chen et.al, 2017). Chen, Gao and Rau (2015) found that users complained about the abundance of the information
conveyed by Danmaku and too much Danmaku leads to a distraction from concentrating on the video content. Moreover, low quality of the comments as well as the look and the feel also enlarge the possibilities to not enjoy the Danmaku for the audiences. Previous studies on the effect of Danmaku were mainly focusing on the benefits of the Danmaku, and the lack of the irritating aspects is a limitation.
Thus, this study hypothesized:
H4. When watching a Danmaku video, viewer’s feel more irritated than when watching the video without Danmaku.
Nationality as a moderator
Although globalization is rapidly developing in this age (McQuail, 2010), the difference of cultural background and media landscape cannot be neglected in media consumption studies. Previous studies on Danmaku were mainly focusing on Danmaku and Chinese young adult users (e.g., Chen et al., 2015). However, different cultures and different languages might affect the level of impact of Danmaku on the viewers (Chen et al., 2017). The lack of the cross-culture perspective is a limitation in Danmaku studies. Thus, the current study also aims to specifically investigate whether there will be differences between Dutch and Chinese young adult in Danmaku viewing.
Firstly, Chinese culture differs to the one of the Netherlands. Different cultures influence the media preferences and media consumptions (Bowman, Jöckel, & Dogruel, 2012). Dutch people have gradually used ‘delayed’ watching and services like Netflix and YouTube to enjoy the audio-visual products (Netherlands, 2020). However, as far as online videos are concerned, Chinese netizens use Chinese online video sites intensively. Netflix and YouTube are not the first choice for Chinese netizens. Moreover, Danmaku video has already been very popular in China and has been adopted by the main Chinese video sites (e.g., Youku.com, IQiyi.com, Bilibili.com etc.). Therefore, for Chinese young adult audiences, Danmaku videos are relatively familiar for them. Based on previous studies, familiarity indicates people’s knowledge of a product or service (Luhmann, 1988). Chen, Lin, Yen and Linn (2011) found that familiarity of a website is a significant factor for the engagement of the website, and high levels of familiarity are positively related to the engagement. Thus, since Chinese audiences are relatively more familiar with the Danmaku video compared to Dutch audiences. Therefore, Danmaku would have a more obvious effect on Chinese young adult audiences than on Dutch young adult audiences.
Second, cyber culture is also of great importance. Danmaku has its special sub-cultural background. In China and Japan, Danmaku video was firstly shared and enjoyed by the sub-culture group who enjoyed the AGC (i.e., animation, comics and game) content the most (Chen et.al, 2015). People share similar interests in this community thus this enhances the feeling of belonging when they watch the Danmaku video. ‘Otaku culture’, which indicates the sub-culture of being alone at home and watching their interests in animation, played an important role in developing the Danmaku community (Li, 2017). Most of the viewers who previously engaged in Danmaku viewing do not have much social interaction in daily life (Zhao, Peng, Tang, & Song, 2017). Thus, Danmaku is a supplement for their social life. Users favor membership and participation, and feel belonging to a specific group (Liu,
Suh & Wagner, 2017). Thus, since Chinese audiences are relatively more familiar with the AGC sub-culture than Dutch audiences, Danmaku would have a more obvious effect on Chinese young adult audiences and on Dutch young adult audiences.
Therefore, the current study hypothesized:
H5. During Danmaku video viewing, Chinese young adult audiences will experience higher level of social presence than Dutch young adult audiences.
H6. During Danmaku video viewing, Chinese young adult audiences will for more entertained than Dutch young adult audiences.
H7. During Danmaku video viewing, Chinese young adult audiences will perceive higher level of information value of the video than Dutch young adult audiences.
Since most of Dutch are unfamiliar with Danmaku and have no habit of watching it. Thus, the tolerance level of the distraction from the abundant comments on the screen can be lower for Chinse audiences, who are more familiar with Danmaku (Chen et al., 2017).
Thus, the current hypothesized:
H8. During Danmaku video viewing, Dutch young adult audiences will feel more irritated than Chinese young adult audiences.
Conceptual model
Methods Participants and sampling
In this study, a convenience sampling method was used. The study recruited the respondents through personal connections (N = 131) and through an online research company website, namely the Prolific (N = 58), resulting in a sample of 189 respondents between 18 and 30 years old, Myears = 22.95, (SD = 3.10). There is a relatively equal gender distribution with 51.9% female and 47.1% male. In addition, the first language distribution is also equal with 52.9% Chinese and 41.7% Dutch. Most of the participants obtained a bachelor’s degree and a high school level education (i.e., 49.2% & 23.8% respectively).
Procedure and design
In order to restore the real day-to-day online video viewing as much as possible, a survey-embedded online experiment was used. Respondents received a link (i.e., Qualtrics) of this experiment through social media or email, where they filled in the consent form and did the experiment. Respondents were informed ahead that their anonymities were
safeguarded and were allowed to end the survey without any reason. The study was approved
Danmaku
Social Presence
H1(+)Entertianed Feelings
H2(+) Percieved Information Value of The VideoH3(+)
Feelings of Irritation
H4(+)
Nationality
by the ethical committee of the Department of Communication Science at University of Amsterdam.
Besides the respondents being reached by personal connections, the Prolific data were collected using the funding of £ 59.48, which was provided by the researcher of the current study to the Prolific website. Based on the database provided by Prolific, this study filtered the targeting respondents by nationality and age (i.e., Dutch people, aged 18 to 30). 59 respondents were randomly selected from the 525 eligible participants to participate in the research. A unique link provided by Prolific was pasted on the last briefing page of the online experiment survey questionnaire. Respondents clicked the link to confirm the participation. In the end, 58 people were approved by the researcher, and they received participation funds (i.e., £0.72 for each) from the Prolific.
A 2x2 factorial design was conducted in this experiment, with Factor Danmaku video as between-subjects variable (2 levels, namely: absent and present) and Factor Nationality as quasi-experimental variable (2 levels, namely: Chinese and Dutch). However, in this study, Chinese participants received Chinese Danmaku whereas Dutch participants received English Danmaku. Language sufficiency and media landscape were the two main concerns of the English Danmaku chosen for Dutch people. Dutch audiences are used to viewing social media comments in English and they usually have a high command of the English language.
In this study, respondents were conditionally assigned into one of the four groups (Danmaku X Chinese, None X Chinese, Danmaku X English, None X English). Respondents with Chinese as their first language were chosen randomly and were assigned to one of the two groups (One with Danmaku in Chinese, N = 50, another without Danmaku，N = 50). Respondents with Dutch as their first language were sent to Element B and were randomly assigned to one of the two groups (One with Danmaku in English, N = 46, another without Danmaku, N = 43). Additionally, the survey was translated into Chinese from English. Thus,
the survey has two language versions (i.e., Chinese & English) for respondents to choose with a question on language sufficiency to finish the survey by the end of the demographic block.
All of the respondents were asked to watch a video with the same content that lasted for 2 minutes and 2 seconds. A fictional scenario description in text on user investigation from a newly built video site called YouBlili was presented to respondents before the video viewing. After the video viewing, the manipulation check question was presented.
Respondents were presented with same questions and demographic block in the following. However, there was one filtered question on frequency of the Danmaku usage, which was only showed to respondents who chose that they knew the item of Danmaku. I emphasized that the video site YouBili was non-existent by the end of the survey. On the last page, participants were thanked for their efforts and they were debriefed.
Experimental materials
In total, 3 videos were designed and edited for this study, using the software Final Cut and Compressor. Video A was the original video without the Danmaku; Video B was the video with the Chinese Danmaku; Video C was the video with the English Danmaku.
A non-conversation animation video called 《玛门》Mammon was chosen from the video site Bilibiil.com. It was posted both on Bilibili and YouTube. The video was free to re-post as set by the uploader, and permission from the uploader was asked before the video editing. The original video lasts for 2 minutes and 45 seconds, with 340 Danmaku in Chinese by the date of 10 December 2019 at 13:00. Among 340 Danmaku comments, I extracted 20% of the total Danmaku (i.e., 65 comments) and translated them into English (See Appendix B).
For all three videos, a school badge icon of UvA was presented at the upper right corner from beginning to the end of the video (See Appendix C). The list of production details in the end of the video was cut and it turned out to be 2 minutes and 2 seconds in total
after editing. However, we have presented the production organization (i.e., CUC Anime) in text before the video block and also thanked the production team in debriefing block.
We have kept the duration and appearance at certain time point of each Danmaku as almost the same as the original Danmaku video, with some of the adjustments to make the English version more adaptable. The font, color and size of Chinese Danmaku and Enlish Danmaku were consistent (i.e., Helvetica Regular, White with RGB (255,255,255), 63.0). The speed, duration time, appearance time point of each pair of Chinese and English Danmaku was consistent. There were 4 lines of Danmaku (i.e., Y axis with 116px, 165px, 221px & 291px), and the position with Y axis was consistent of each Chinese and English Danmaku pair.
Measurements
Familiarity. Familiarity consists of the familiarity level of the video Mammon and the familiarity level of the Danmaku. To measure the familiarity level of the video Mammon, one item was designed (i.e., Have you ever watched this Video Mammon before?). Yes, No and Don’t remember were the answer scales. Category Yes was coded as 1, while other answer categories were coded as 0. The outcome shows that none of the respondents have seen the video Mammon before. Thus, the familiarity level of the video Mammon will not be included in the analysis. To measure the familiarity level of Danmaku, two items were designed. One fixed item (i.e., Have you heard of the term Danmaku /Bullet screen/ 弹幕 before?) was shown to every respondent. Yes and No were the answer scales (i.e., 61.4%Yes, 38.6% No). One filtered item (i.e., How often have you watched Danmaku /Bullet screen/ 弹 幕 videos before?) was shown to respondents who reported they knew Danmaku before. Respondents rated each item on a 5-point Likert scale ranging from 1= Never to 5=Every time (M = 2.91, SD =1.01).
Perceived Social Presence. Respondents indicated their level of social presence using the five items (See Appendix D) of Social Presence Scale adapted from Gefen,
Karahanna and Straub (2003). For example, “I felt others were watching at the same time as I did.” Respondents rated each item on a 5-point Likert scale ranging from 1= strongly
disagree to 5=strongly agree. In total, the factor explained 68.2% of the variance in the 5 items (α = .88, M = 2.81, SD = 1.08). A principal axis factor analysis with Direct Oblimin rotation was conducted.
Entertained Feelings. 5 items adapted from Voss, Spangenberg and Grohmann (2003) measured ‘Entertained Feelings’ (See Appendix D). Respondents rated each item (e.g., To what extend do you feel happy?) on a 5-point Likert scale ranging from 1= not at all to 5=To a very great extent. Two factors were extracted based on the Eigenvalue‐criterion (bigger than 1: eigenvalue factor 1 is 2.67; and factor 2 is 1.25). In total, the factors explained 76.41% of the variance in 5 items, with factor one accounting for 51.48% of the variance explained, and factor two added 24.93% of the explained variance. Three items measured Factor 1, stated hedonic mood (α = .76, M = 2.19, SD = 0.87). Two items measured Factor 2, stated hedonic value of the video (α = .77, M = 2.85, SD = 0.85). According to the H2, the current study only included factor 1, which indicated the hedonic mood, in the following analysis. A principal axis factor analysis with Direct Oblimin rotation was conducted.
Perceived Information Value of the Video. To measure the perceived information value of the video, the current study adapted five items (See Appendix D) from the Voss, Spangenberg and Grohmann (2003). Respondents rated each item (e.g., To what extent do you think this video is useful?) on a 5-point Likert scale ranging from 1= not at all to 5=To a very great extent. In total, the factor explained 79.7% of the variance in the 5 items (α = .94, M = 2.76, SD = 0.94). A principal axis factor analysis with Direct Oblimin rotation was conducted.
Irritated Feelings. The Irritation Scale consists of four items (See Appendix D), adapting from Chen, Gao and Rau (2017)’s focus group results on Danmaku (e.g., I felt the information conveyed by the video make me feel tired). Respondents rated each item on a 5-point Likert-type scale ranging from 1= strongly disagree to 5=strongly agree. In total, the factor explained 65.7% of the variance in the 4 items (α = .82, M = 2.61, SD = 1.05). A principal axis factor analysis with Direct Oblimin rotation was conducted.
Manipulation check
One manipulation check question was used to test on the presence and absence of Danmaku. Respondents indicated whether they have seen the Danmaku or not; the answer categories were (a) Yes or (b) No or (c) Do not remember. A chi-square test confirmed successful manipulation of Danmaku presence (c2(2) = 174.55, p < .001).
Analytical Strategy
The level of the dependent variable perceived social presence, perceived
entertainment feelings, perceived information value of the video and irritated feeling of the video are interval, and the independent variable and moderation variable are at nominal level. To test all of the hypotheses, the two-way ANOVA is suitable. Thus, 4 two-way ANOVA was conducted separately.
Results Descriptions
Most of the Dutch respondents (81%) reported they have never heard of Danmaku before, while only 1% of the Chinese respondents reported they have never heard of the Danmaku before. Among the respondents (i.e., Chinese & Dutch) who reported they have known about Danmaku before, 69% of them watch the Danmaku at least on the sometimes level. Specifically, 77.8 % of the Chinese respondents reported they watch the Danmaku at least sometimes, and no Dutch respondents watch the Danmaku more than sometimes.
Perceived Social Presence
To investigate the effect of Danmaku viewing on perceived social presence and to test H1 and H5, a two-way analysis of variance was carried out with Danmaku presence and nationality as independent variable, and social presence as dependent variable.
The result showed a statistically significant main effect of Danmaku, F (1, 185) = 18.11, p < .001, 𝜂2 = 0.07. Young adult audiences who viewed the video with Danmaku perceived higher level of social presence than those who didn't view the Danmaku (Mdif =
0.53). The main effect of nationality was also statistically significant, F (1, 185) = 56.71, p < .001, 𝜂2 = 0.21. This indicates that on general, Dutch young adult audiences (M = 2.29, SD = 1.11) perceived lower social presence than Chinese young adult audiences (M = 3.28, SD = 0.82). The interaction effect between Danmaku use and nationality was statistically
significant (F (1, 185) = 7.17, p = .004, 𝜂2 = 0.03), suggesting that nationality moderates the effect of Danmaku on social presence. Specifically, Chinese respondents reported a higher level of perceived social presence (M = 3.36, SD = 0.84) than Dutch respondents (M = 2.75, SD = 1.10) during Danmaku video viewing session. The overall model explained 30% (R2=.30) of the variance in the perceived social presence when viewing the online video. All the means can be found in table 1. The differences between Dutch and Chinese young adult audiences are visible in figure 1.
It should be noted that the assumption of equal variances in the population has been violated, Levene's F (3, 185) = 4.49, p = .005.
Table 1. Descriptive Statistics for Perceived Social Presence
Danmaku exposure Nationality Mean Std Deviation N
Presence Chinese 3.36 0.84 50 Dutch 2.75 1.10 46 None Chinese 3.19 0.79 50 Dutch 1.80 0.91 43 Total Chinese 2.29 1.11 89 Dutch 3.28 0.82 100
Figure1: Means of Perceived Social Presence
Feelings of being entertained
To investigate the effect of Danmaku viewing on feelings of being entertained and to test H2 and H6, a two-way analysis of variance was carried out with Danmaku presence and nationality as independent variable, and feelings of being entertained as the dependent variable.
I did not find a significant effect of exposure to the Danmaku on the respondents’ feeling of being entertained (F (1, 185) = 2.57, p = .111) or interaction effect of Danmaku presence and nationality (F (1, 185) = 0.58, p = .447). However, the effect of nationality was significant (F (1, 185) = 24.86, p < .001, 𝜂2 = 0.12). This indicates that in general, Dutch respondents reported a higher level of feeling being entertained (M = 2.51, SD = 0.79) than Chinese respondents (M = 1.92, SD = 0.84). The overall model explained 13% of the variance in perceiving hedonic feelings when viewing the online video.
Perceived Information Value of the Video 1,50
2,00 2,50 3,00 3,50
Danmaku Absent Danmaku Present
To investigate the effect of Danmaku viewing on perceived information value of the video and to test H3 and H7, a two-way analysis of variance was carried out with Danmaku presence and nationality as independent variable, and perceived information value of the video as dependent variable.
We did not find a significant effect of exposure to the Danmaku on the respondents’ perceived information value of the video (F (1, 185) = 0.50, p = .823) or interaction effect of Danmaku presence and nationality (F (1, 185) = 0.02, p = .900). However, the effect of nationality was significant (F (1, 185) = 52.08, p < .001, 𝜂2 = 0.22). Specifically, Chinese respondents reported a higher level of perceived information value of the video (M = 3.18, SD = 0.82) compared to Dutch respondents (M = 2.30, SD = 0.85). The overall model
explained 22% of the variance in perceiving hedonic feelings when viewing the online video. Feelings of Irritation
To investigate the effect of Danmaku viewing on feelings of irritation and to test H4 and H8, a two-way analysis of variance was carried out with Danmaku presence and
nationality as independent variable, and feelings of irritation as dependent variable. The results showed a statistically significant main effect of Danmaku, F (1, 185) = 59.86, p < .001, 𝜂2 = 0.21. Young adult audiences who viewed the video with Danmaku perceived higher level of irritations than those who didn't view the Danmaku. The effect of nationality is statistically significant, F (1, 185) = 4.36, p = .038, 𝜂2 = 0.01. This indicates that in general, Dutch young adult audiences (M = 2.76, SD = 1.29) felt more irritated than Chinese young adult audiences (M = 2.48, SD = 0.75).The interaction between Danmaku use and nationality was statistically significant (F (1, 185) = 46.78, p <.001, 𝜂2 = 0.16),
suggesting that nationality moderates the effect of Danmaku on irritations. Specifically, Dutch respondents reported a higher level of feeling irritated (M = 3.62, SD = 1.00) than Chinese respondents (M = 2.53, SD = 0.79) during Danmaku video viewing session. The
overall model explained 36% (R2=.36) of the variance in perceiving social presence when viewing the online video. For all means, refer to table 2. The differences between Dutch and Chinese young adult audiences are presented in figure 2.
Table 2. Descriptive Statistics for Feelings of Irritation
Danmaku exposure Nationality Mean Std Deviation N
Presence Chinese 2.53 0.79 50 Dutch 3.63 0.10 46 None Chinese 2.42 0.71 50 Dutch 1.84 0.85 43 Total Chinese 2.48 0.75 89 Dutch 2.76 1.29 100
Figure2: Means of Feelings of Irritation
Discussion
Danmaku is a relatively new form of online comments in online video viewing. It has been a success in east Asia, especially in China and Japan. However, it is relatively unknown to westerners including Dutch people. Danmaku has drawn some attention in academia on motivations. However only few studies have explored Danmaku’s effect on young adult
1,50 2,00 2,50 3,00 3,50 4,00
Damaku Absent Danmaku Present
audiences (Fang et.al, 2018). This study aimed to better understand what viewing experiences Danmaku videos can provide to young adults, as well as the differences between Dutch and Chinese young adults viewers.
The biggest feature of the Danmaku is its social feature. Compared to traditional online video comments (i.e., forum comments), the present study found that Danmaku can provide a higher level of perceived social presence for the viewers. This is in line with previous studies that observed that people exposed to Danmaku obtained a feeling of
belonging and a sense of being accompanied (Chen et.al, 2015;). This social feature creates a co-viewing experience for both Dutch and Chinese young adult audiences. Moreover, the study extends previous research that found the effect of feeling social presence during Danmaku video viewing is moderated by the nationality. Comparing the two nationalities (i.e., Dutch and Chinese), the current study found that overall, Chinese young adult audiences experienced more social presence than Dutch young adult audiences either with or without the Danmaku. However, Dutch young adult audiences experienced a more dramatic rise of social presence after the Danmaku viewing.
Although previous studies have shown that Chinese young adult audiences watch Danmaku videos for the feeling of being more entertained (Chen et.al, 2015; Chen et.al, 2017; Djamasbi et.al, 2016). In the current study, neither Chinese young adult audiences nor Dutch young adult audiences showed that they felt more entertained after Danmaku video viewing. A similar phenomenon can be seen in the effect of perceived information value of the video as well. Although it was expected that Danmaku as a text comment feature can provide young adult audiences with a sense of obtaining more information during video watching, this expectation was not supported in the current study. These findings are
controversial compared to previous studies (i.e., people use Danmaku to feel entertained and seek information, Chen et.al, 2015; Chen et.al, 2017). Individual differences are a possible
reason to explain why the findings in the current study are not in line with the previous ones. Previous studies were trying to figure out the motivations of Danmaku use, thus people who like to use the Danmaku provided positive reasons and effects for using the Danmaku, while people who did not enjoy the Danmaku reported more negative aspects. In the current study, many of the respondents were unfamiliar with Danmaku and did not have the habit of watching the Danmaku. This would also affect the acceptance and enjoyment of the Danmaku viewing.
However, an interesting finding needs to be pointed out: overall, Dutch young adult audiences enjoyed the video more than Chinese young adult audiences. The possible reason for this difference between Chinese and Dutch young adult audiences might be the genre of the video. The anime the study has chosen (i.e., Mammon《玛门》) is a non-dialogue animation with a Christian background. It is culturally closer to the Dutch, which may be the possible reason for why Dutch young adult audiences liked this video more.
Additionally, it turned out to be opposite in perceived information value of the video: Chinese young adult respondents reported they had obtained more information from video viewing than Dutch young adult audiences. Anime with religious features are relatively new to young Chinese audiences, this knowledge gap created a room for Chinese young adult audiences to learn new things from video viewing.
Previous studies also addressed some irritating features of the Danmaku (i.e.,
information abundance, aesthetic features etc.). It was expected that during Danmaku video viewing, the text on the screen would have an influence on making viewers feel negative, and the current study met that expectation. The findings show that people feel more irritated after Danmaku viewing. This result is in line with the previous studies that Danmaku itself can also have some negative factors (e.g., Chen et.al, 2015). These negative factors may also decrease the entertainment experiences of the viewers. The effect of feeling irritated during
Danmaku video viewing is moderated by nationality. Compared to Chinese young adult audiences, Dutch young adult audiences felt more irritated after the Danmaku viewing. Only 1% of the Chinese young adult respondents have not heard about the item of Danmaku while 81% of the Dutch young adult repondents have known Danmaku before, which indicates most of the Dutch people have not experienced the Danmaku before and this is the possible reason for them to feel irritated. Dutch young adult audiences are new to Danmaku and not familiar with this new way of online video viewing, which enlarged their negative
perceptions of the Danmaku.
Limitations and Suggestion for Future Research
The study has some limitations that need to be considered. First, individual
differences were not considered in this experimental study. In Chen, Gao and Rau (2017)’s study, they found that introverted young men who were open to new experiences are more likely to enjoy the Danmaku. People's level of acceptance and familiarity with Danmaku should also be controlled. Second, the current study only included one specific genre of the video. Previous studies have addressed that Danmaku is popular most on UGC (i.e., User-Generated -Content) websites (e.g., Chen et.al, 2015; Chen et.al, 2017; Lan, Cha, Cho & Hwang, 2019), and audiences enjoy the Danmaku more when the video content contains entertainment. In the current study, I only considered the video highly related to
entertainment content. Third, only two nationalities were included in the experiment and sub-culture groups were not taken into account. For example, Danmaku video sites were firstly popular among the AGC (i.e., Animation, Comic, Game) subculture groups, and AGC groups continually became the largest community of Danmaku (Peng, Zhao, & Teo, 2016), it might be different if the study narrowed down the population to Dutch and Chinese young adults who enjoy the AGC. Future experimental research may therefore manipulate not only the
Danmaku presence but also the genre of the videos, subculture influence, as well as take individual differences into consideration.
Conclusion
This study affirmed that Danmaku increases the feelings of social presence to both Chinese and Dutch young adult audiences. Although Chinese audiences perceived higher levels of social presence, the increase of feeling socially present when exposed to a Danmaku video was higher for Dutch young adult audiences. Despite these positive outcomes,
Danmaku also has irritating features for the audiences and Dutch young adult audiences were more aware of the Danmaku’s negative aspects. The results support the ongoing research on the effects of online video comments, which has consistently shown that Danmaku improves the engagement of the videos. This means that researchers who are interested in
understanding the role of Danmaku among young adults will need to investigate how Danmaku would affect the viewing experiences.
References
Aragon, S., & Aragon, S. (2003). Creating social presence in online environments. New Directions for Adult and Continuing Education, 2003(100), 57–68.
https://doi.org/10.1002/ace.119
Arnett, J. (2004): A longer road to adulthood. In: Emerging adulthood: The winding road from late teens through the twenties (chapter 1). Oxford: Oxford University Press. Bondad-Brown, B., Rice, R., & Pearce, K. (2012). Influences on TV Viewing and Online User-shared Video Use: Demographics, Generations, Contextual Age, Media Use, Motivations, and Audience Activity. Journal of Broadcasting & Electronic Media, 56(4), 471–493. https://doi.org/10.1080/08838151.2012.732139
Bowman, N., Jöckel. S., & Dogruel, L. (2012). A question of morality? The influence of moral salience and nationality on media preferences. Communications - The
European Journal of Communication Research, 37(4), 345–369. https://doi.org/10.1515/commun-2012-0020
Chen, Y., Gao, Q., & Rau, P. (2015). Understanding gratifications of watching danmaku videos – videos with overlaid comments. 9180, 153–163. https://doi.org/10.1007/978-3-319-20907-4_14
Chen, Y., Gao, Q., & Rau, P. (2017). Watching a Movie Alone yet Together: Understanding Reasons for Watching Danmaku Videos. International Journal of Human-Computer Interaction, 33(9), 731–743. https://doi.org/10.1080/10447318.2017.1282187 Chen, J., Lin, C., Yen, D., & Linn, K. (2011). The interaction effects of familiarity, breadth
and media usage on web browsing experience. Computers in Human Behavior, 27(6), 2141–2152. https://doi.org/10.1016/j.chb.2011.06.008
Chorianopoulos, K., & Lekakos, G. (2008). Introduction to Social TV: Enhancing the Shared Experience with Interactive TV. International Journal of Human-Computer
Interaction, 24(2), 113–120. https://doi.org/10.1080/10447310701821574 Coyne, S., Padilla-Walker, L.M., & Howard, E. (2013). Emerging in a digital world: A
decade review of media use, effects, and gratifications in emerging adulthood. Emerging Adulthood, 1, 125-317.
Davis Mersey, R., Malthouse, E., & Calder, B. (2010). Engagement with Online Media. Journal of Media Business Studies, 7(2), 39–56.
https://doi.org/10.1080/16522354.2010.11073506
Djamasbi, S., Hall-Phillips, A., Liu, Z., Li, W., &Bian, J. (2016). Social Viewing, Bullet Screen, & User Experience: A First Look. 2016 49th Hawaii International Conference on System Sciences, 648-657. doi: 10.1109/HICSS.2016.86 Fang, J., Chen, L., Wen, C., & Prybutok, V. (2018). Co-viewing Experience in Video
Electronic Commerce, 22(3), 446–476.
https://doi.org/10.1080/10864415.2018.1462929
Gefen, D., Karahanna, E., & Straub, D. (2003). Trust and TAM in online shopping: An integrated model. Mis Quarterly, 27(1), 51–90. https://doi.org/10.2307/30036519 Haridakis, P., & Hanson, G. (2009). Social Interaction and Co-Viewing With YouTube:
Blending Mass Communication Reception and Social Connection. Journal of Broadcasting & Electronic Media, 53(2), 317–335.
https://doi.org/10.1080/08838150902908270
Howard, R. (1966). Information Value Theory. IEEE Transactions on Systems Science and Cybernetics, 2(1), 22–26. https://doi.org/10.1109/TSSC.1966.300074
Jing Leng, Jiayu Zhu, Xiaoting Wang, & Xiaoqing Gu. (2016). Identifying the Potential of Danmaku Video from Eye Gaze Data. 2016 IEEE 16th International Conference on Advanced Learning Technologies (ICALT), 288–292.
https://doi.org/10.1109/ICALT.2016.155
Li, J. (2017). The Interface Affect of a Contact Zone: Danmaku on Video-Streaming
Platforms. Asiascape: Digital Asia, 4(3), 233–256. https://doi.org/10.1163/22142312-12340079
Lin, X., Huang, M., & Cordie, L. (2018). An exploratory study: using Danmaku in online video-based lectures. Educational Media International, 55(3), 273–286.
https://doi.org/10.1080/09523987.2018.1512447
Liu, L., Suh, A., & Wagner, C. (2016). Watching online videos interactively: the impact of media capabilities in Chinese Danmaku video sites. Chinese Journal of
Liu, L., Suh, A., & Wagner, C. (2017). Who is with you? Integrating a play experience into online video watching via danmaku technology. 10272, 63–73.
https://doi.org/10.1007/978-3-319-58077-7_6
Martin, E. (2018). The state of online video. EContent, 41(1), 19–21.
McQuail, D. (2010). McQuail’s mass communication theory (6th ed). London: SAGE. Netherlands. (2020). Retrieved 30 January 2020, from
https://medialandscapes.org/country/netherlands
Oh, C., Bailenson, J., & Welch, G. (2018). A Systematic Review of Social Presence: Definition, Antecedents, and Implications. Frontiers in Robotics And AI, 5. doi: 10.3389/frobt.2018.00114
Peng, X.; Zhao, Y., & Teo, H., (2016). Uunderstanding Young People’s Use of Danmaku Websites: The Effect of Perceived Coolness And Subcultural Identi. PACIS 2016 Proceedings. 252. http://aisel.aisnet.org/pacis2016/252
Purcell, K. (2010). Retrieved 7 January 2020, from
http://uploadi.www.ris.org/editor/1276126693PIP-The-State-of-Online-Video.pdf Siersdorfer, S., Chelaru, S., Nejdl, W., & San Pedro, J. (2010). How useful are your
comments?: analyzing and predicting youtube comments and comment ratings. Proceedings of the 19th International Conference on World Wide Web, 891–900. https://doi.org/10.1145/1772690.1772781
Team, U. (2020). Aurora Big Data: 2017 Bilibili Research Report. Retrieved 27 January 2020, from https://www.useit.com.cn/thread-16775-1-1.html [In Chinese]
Voss, K., Spangenberg, E., & Grohmann, B. (2003). Measuring the Hedonic and Utilitarian Dimensions of Consumer Attitude. Journal of Marketing Research, 40(3), 310–320. https://doi.org/10.1509/jmkr.40.3.310.19238
Wu, Q., Sang, Y., Zhang, S., & Huang, Y. (2018). Danmaku vs. Forum Comments: Understanding User Participation and Knowledge Sharing in Online Videos.
Proceedings of the 2018 ACM Conference on Supporting Groupwork, 209–218. https://doi.org/10.1145/3148330.3148344
Yan, L., Cha, N., Cho, H., & Hwang, J. (2019). Video Diffusion in User-generated Content Website: An empirical analysis of Bilibili. 2019 21st International Conference on Advanced Communication Technology (ICACT), 2019-, 81–84.
https://doi.org/10.23919/ICACT.2019.8701897
Zhang, Y., Qian, A., Pi, Z., & Yang, J. (2019). Danmaku Related to Video Content Facilitates Learning. Journal of Educational Technology Systems, 47(3), 359–372. https://doi.org/10.1177/0047239518811933
Zhao, Y., Peng, X., Tang, J., & Song, S. (2017). Understanding Young People’s We-Intention to Contribute in Danmaku Websites: Motivational, Social, and Subculture Influence. iConference 2017 Proceedings. https://doi.org/10.9776/17207
Appendices Appendix A
Appendix B Table B. Danmaku content table
Danmaku Content (Chinese vs. English) Category
1 我们都生活在阴沟里，但总有人仰望星空。这才是讽刺的意义。
We all live in gutters, but someone looks up at stars. That's the Irony.
S
2 玛门=钱？
Mammon = Money?
S
3 弹幕还没看的人多？
Danmaku comments are less than the people now watching?
C
4 玛门意为象征之恶魔
Mammon means the demon
I
5 不信耶稣信玛门
believe in Mammon not Jesus
S
6 玛门一词最早在圣经上记载，多指偶像和钱财
Mammon was first mentioned in the Bible, means idols and money
I
7 玛尼
Ma god
E
8 这里科普一下，玛门作为七宗罪里的恶魔之一，贪婪之罪
Attention here, Mammon belongs to seven deadly sins, is the greed
I
9 玛门的意思是金钱
Mammon means money
I
10 画风赞 amazing style
E
11 欲望与追求才是人的本性，贪婪是人私心太重造成的 Desire& pursuit=human nature, greed caused by selfishness
S
12 火车来了，呜呜呜呜呜呜 Train sound, chug and chuff
E 13 哎呀妈呀 Oh my goodness E 14 孤岛惊魂 2 Far Cry 2 S 15 这画风…… emm, this style
E
16 捡空投的我
It is me picked up the airdrop
17 抢空投？ Grab an airdrop?
E
18 这音乐与画面一点都不配 music doesn't match video at all
S
19 抢空投啦 grab airdrop!
E
20 捡空投的你
It is you picked up the airdrop
C
21 没人吗
anybody online?
C
22 有这身手还住这种破房子？
People who have such skill living in such shabby house?
S
23 捡空投的笑死我 Lol, airdrop
C
24 在这个沙漠里，来点水，也比砖石有吸引力吧
In this desert, I think water is more attractive than this stone?
S 25 哇塞，一块石头
Wow, a big stone
E 26 眼神空洞 Eyes hollow S 27 萌 cute E 28 玛门，古迦勒底语
Mammon, ancient Chaldean
I 29 哎哟，两个妹子
Ouch, two ladies
S 30 哈哈哈哈哈
hahahahah
E 31 原始部落的人要钻石有毛用
people from primitive tribe need a diamond?
S 32 抢宝石
Fight for gems
S 33 周围都是沙漠，一块钻石还没一瓶水有用
desert all around, water is better than a diamond
S 34 一块石头有什么用 Useless stone S 35 666 So sick E 36 没穿 No underwear E 37 没穿的怕不是瞎了，是肉色的
Are you blind? The underwear is flesh-colored
38 邻里关系极差
Neighborhood is extremely poor
E 39 你们这群变态……
Your group of perverts
C 40 不能把宝石卖了再平分吗？
Can't they sell the gem and share the money?
S 41 换成黄金更保值，钻石的话不适合
replace it with gold. more valuable than diamonds.
S 42 让巨龙吞噬你们
Let the dragon devour you
C 43 这个故事告诉我们邻居关系相处不友好，一见面就是这个结果
if neighbors get along badly, then this is the result when they meet
S 44 女人，喜欢闪亮亮的宝石
Woman, like shiny stones
E 45 感觉中间放桶水或者井更有意思
more interesting if replace the gem with water
S 46 原来是个动作片
Turned out to be an action movie
S 47 23333
Lol E
48 我以为撞在一起
I thought they bumped together
S 49 双杀
Double kill
E 50 人是贪心的
People are greedy
S 51 这种地方钻石只能当桌子用
In this kind of place, diamond can only be used as a table
E 52 就算抢到了，又能怎样
Even if you won gem, and then?
S 53 百年孤独？
A century of loneliness?
S 54 人类进化史
History of human evolution
S 55 玛门是“贪婪”之原罪的化身一类的人
Mammon = people born of greed
I 56 灭霸？ Thanos S 57 没看懂 Didn't get it S 58 因贪婪而争斗的结果只能两败俱伤，这更会使欲望本身变得诱人
a fight over greed is to lose both, which makes desire itself attractive
S 59 讽刺个屁啊
Ironic?
C
anyone remember the grass in the beginning? 61 真好吃 So tasty S 62 人的欲望？ Human desire? S 63 这株小草才是荒漠中最大的财富
This grass is the greatest wealth in the desert
S 64 我感觉真幼稚 I feel so naive S 65 草：MVP Grass: MVP E 1.Self-expression -S 2. Company-C 3. Information- I 4. Entertainment-E
Appendix C
Appendix D Table D. Questionnaire
Familiarity
F-1: Have you ever watched this Video Mammon before?
F-2: Have you heard of the term Danmaku /Bullet screen/ 弹幕 before?
F-3: How often have you watched Danmaku /Bullet screen/ 弹幕 videos before? Entertainment/Hedonic
EH-1: To what extend do you think this video is interesting? EH-2: To what extend do you think this video is amusing? EH-3: To what extend do you feel relaxed?
EH-4: To what extend do you feel happy? EH-5: To what extend do you enjoy the video? Information Value/Utilitarian
I-1: To what extent do you think this video is useful? I-2: To what extent do you think this video is wise? I-3: To what extent do you think this video is valuable? I-4: To what extent do you think this video is beneficial? I-5: To what extent do you think this video is helpful? Social Presence
SP-1: I felt others were watching at the same time as I did.
SP-2: I felt I wanted to share my opinions of the video when I was watching. SP-3: I felt I was part of a larger audience when I was watching the video. Sp-4: I felt connected to others when I was watching the video.
Irritation
IR-1: I felt the whole video contains too much information.
IR-2: I felt the information conveyed by the video make me feel tired. IR-3: I could not focus on watching the video itself.
