A Study on the Effects of Fandom's Digital Media Use on Social Participation : Focused on the Mediating Effect of the Identification

팬덤의 디지털 미디어 이용이 사회참여에 미치는 영향 연구 : 동일시의 매개 효과를 중심으로

Abstract

In this paper, we attempt to empirically explain the process by which fandom performs social participation using digital media. To do this, we utilize the concept of identification and the Situational Theory of Publics in a theoretical framework, and study ARMY, BTS fandom. We confirmed a positive association between communication behavior and social participation, and verified a mediating effect of the identification as well. Interestingly, horizontal identification, a fan-to-star friendly relationship, had a greater effect on moving toward social participation than did vertical identification, a fan's blind worship of a star. This study has significance in that it has figured out the relationship of personal feelings to public participation by mobilizing achievements in various research fields.

본 연구는 디지털 미디어를 이용하는 팬덤이 사회적 참여를 수행하는 과정을 실증적으로 해명하는 데 목적이 있다. 이론적인 틀로 동일시 개념을 새롭게 해석하고 그루닉의 공중 상황 이론을 활용한다. 이를 위해 온라인이나 소셜 미디어를 활용하면서 방탄 소년단(BTS)의 가치를 실제 행동으로 실천하고 있는 팬덤, 아미를 대상으로 설문 조사를 실시하였다. 연구결과, 아미의 디지털 매개 커뮤니케이션 행위와 사회참여 사이의 긍정적인 관련성을 검증하였고, 아미가 방탄소년단에 대하여 경험하는 동일시가 매개 역할을 한다는 사실 역시 확인하였다. 특히, 개인이 팬 대상을 친구나 동료로 받아들이는 수평적 동일시는 팬 대상을 스타로 여기는 수직적 동일시보다 사회참여로 나아가는 데 더 큰 영향을 미쳤다. 본 연구는 디지털 미디어, 미디어 수용자, 팬덤 그리고 PR 등 다양한 연구 분야들의 성과들을 동원하여 개인의 미디어 이용과 사적 감정, 공적 행동 사이의 관련성을 밝혔다는 데 그 의미가 있다.

I. Introduction

In the era of Corona-19 virus, media is more important than ever to modern people. Because of the social environment in which physical contact is socially taboo, people use online and SNS to chat with friends, buy necessary items, and work inside the house. Media based on digital technology has become an important factor in determining the personal lives and social roles of modern people, which is why we should pay attention to fandom, people who had used the digital media effectively before the virus broke out. Fandom experiences fan targets through the media and further uses the media actively[1]. Fans based on passionate personal feelings about a particular object are using online and social media to maintain their identity and exert social influence. What's interesting about fan activism is that fans' private interests can easily and actively be linked to public interests[2]. According to K-pop fandom studies, fandom has attempted to make a social change from private collective practice, such as raising the ranking of K-pop idol singers' online real-time searches or providing food trucks for their star-studded programs, to public collective actions, such as signing campaigns to abolish the music pre-review system, and demanding correction of unfair long-term contracts of singers[3]. In this study we aim to empirically identify the process of the participatory behavior of the fandom. In order to figure out how the personal feelings formed through the digital media lead to public collective action, we newly interpret the identification.

Online fandom, for example, functions as a virtual community around the identification[4]. Fans using Facebook have also shown interesting results, in that their passive identification experience is more effective than is the active one[5]. In particular, BTS' fan club, ARMY, forms a friend-like relationship by experiencing the identification with BTS[6]. We will explore the identification in two dimensions for the study. The vertical identification, which thinks of a star as an ideal to model, and horizontal identification, which considers a star as an intimate friend, will provide clues to gaining a concrete grasp of the process of fandom's social participation in the digital era. Another feature of this study is that we use the Situational Theory of Publics in the PR field[7]. Whereas Grunig assumed the communication actions of individuals in the formation of the public on a particular issue or situation as a dependent variable[7], we assume them as an independent variable, because individuals who frequently use digital media are at the center of the participatory culture while easily experiencing a mixed culture of content penetrating between existing media and new media[2]. This study targets BTS’s digital fandom, ARMY. The case of ARMY, which uses online or social media to practice the value that BTS conveys to the world[6], will be a good example of how a personal communication action using online and SNS leads to social participation through the medium of the identification.

Ⅱ. Theoretical Background

1. Digital Fandom & Comm. Behavior

Using the digital technology, individuals have been reborn as actors who carry out creative activities through communication actions. They can expand their territory indefinitely and easily find people with common interests[4]. The segmented individual interests and tastes are linked around something in digital space[8]. As fans participated in the community, they experienced a media environment that further strengthened solidarity with their fans[4].

Fans interpret media text, actively express their thoughts, and participate in the community[7]. They consume, produce, and share information through the media. To explain the relationship between a fan and media, Fisk divides fans' productivity into semantic, enunciative, and textual[9]. “Semiotic productivity” refers to the production of meaning at the personal level that understands meaning through media text, “enunciative productivity” corresponds to the production of meaning at the community level through face-to-face communication, and “textual productivity” can be understood as sharing the text and the meaning at the community level, both offline and online[9]. In other words, semiotic productivity is the personal information pursuit that actively searches for and produces the meaning of information, whereas enunciative productivity and textual productivity refer to the collective act of expressing and sharing information. Thus, both enunciative productivity and textual productivity can be summed up as information sharing, in that they share information within a community.

Grunig elaborated on the communication behavior of individuals in the formation of the public on specific issues or situations through the Situational Theory of Publics[7]. Individual communication behaviors were divided into information processing activities, in which individuals passively experience information, and information-seeking activities, in which individuals actively search for information[7].

In the paper, we use the Situational Theory of Publics[7] and the concept of fan productivity[9] to describe the communication behavior of fans in the digital media environment. First of all, we present Grunig's proposed acts of information processing and information pursuit as two types of individual communication acts for the study. Fiske's two concepts, enunciative productivity and textual productivity, are named as information-sharing acts, which are added as the third type of individual communication acts.

Specifically, information processing refers to a passive experience at a personal level in which an individual reads or watches given information. In digital virtual space, individuals are exposed to much information; so they can easily experience it at any time[7]. Individuals also experience social effects through simple actions of reading or viewing other people's writings or content[10]. Modern people who passively experience information are sometimes called a reading public, because they reveal their existence and social influence by the number of Internet views or searches[10].

On the other hand, information seeking refers to the meaning produced by an individual who understands, interprets, and evaluates information while actively visiting a specific community or searching for information. Fans who use online or social media find images and writings related to their favorite stars and engage in communication activities that interpret or evaluate their meanings personally[4].

Finally, information sharing is characterized by interaction between members within the community. It refers to the production of information at the online and off-line community level that posts writings and images, or exchanges opinions on certain issues, and shares information with members. Especially, the modern public reveals itself by actively participating in the discourse composition online[10]. Fans engage in communication within the community through routine and repetitive actions, such as clicking 'like' with members[6].

2. Identification

Four main definitions of identification can be found in the existing literature. It can be divided into the emotional state for a fan object, cognitive judgment and attitude, sharing of a goal and a value, and imitation[11]. The various definitions of the identification deal with immersion, intimacy, motivation, and behavior. Looking at how its different definitions are related is a prerequisite for figuring out its concept. Approaching it as a series of processes provides a clue to explaining an individual's behavior[11].

The process of identification begins with the formation of a favorable attitude toward the object, allowing individuals to experience emotional immersion. Specific goals or values are presented in order for attitudes formed here to lead to action[12]. Kelman emphasizes that people actually believe in values shared with objects in the identification[13]. From the perspective of action performance, the general attitude toward an object predicts specific actions, and actions become a means of achieving goals and realizing values [14]. Thus, attitudes formed toward an object share goals or values and act in a specific way to realize them[12][13]. Eventually, the identification process begins with an attitude toward an object and ends with inducing action.

A person who attempts to imitate behavior through identification equates with a role model[15]. However, Bandura did not fully consider the interaction that occurs actively between the imitator and the role model[16]. Ironically, criticism of the identification process sparked interest in a new dimension of identification.

In particular, to explain the audience response to the media experience, celebrity suicide studies specifically presented a new dimension of the identification as vertical and horizontal[17]. The vertical identification refers to the existing concept of identification, equating oneself with someone who is superior or whom one wants to resemble[17]. People tend to identify with people who are well-known and have special abilities.

On the other hand, the horizontal identification, a new dimension of identification, means equating oneself with someone with similar characteristics[17]. The greater the demographic similarity between the object and the individual, the greater the imitative behavior of the individual[18]. Depending on the perception of the social status of the object, the identification is divided into the vertical identification with the object of ascension and the horizontal identification with the object of equality[17].

Vertical and horizontal identification presented in suicide studies are not mutually exclusive[17]. If one's favorite star has factors similar to oneself, one experiences both vertical and horizontal identification with the star. In the Marilyn Monroe Suicide Study[19], male fans of Marilyn Monroe showed only vertical identification with her suicide, whereas female fans experienced both vertical and horizontal identification. News report on a suicide committed by a celebrity similar in age also made it easy to imitate suicide behavior, allowing the audience to experience two types of identification simultaneously[17]. The effect of identification becomes stronger when experienced the two at the same time[17].

3. Comm. Behavior & Identification

Communication behavior of individuals using the media is phased out. After the individual has passively experienced various types of information (information processing), he or she proactively searches for information of interest to understand its meaning (informational pursuit), and then exchanges information with members of the community, producing other information (information sharing).

Brown and his colleagues' study, which observed changes in British people's attitudes and behavior after the media coverage of Princess Diana's death, provides clues to the relationship between communication behavior and identification[20]. Those who first encountered her sudden death through newspapers and television (information processing) identified themselves with her death, then actively sought out articles and news related to her death (information seeking)[20]. The act of processing information, deriving dense communication behavior, stimulates another act of information pursuit. In the process, people shared news about Princess Diana's death with people, exchanging related rumors and news (information sharing)[20]. Based on the experience of the identification, information processing spread to information seeking and to information sharing offline and online. The three communication behaviors are linked to each other rather than being independent actions. For celebrities such as Princess Diana, people experienced vertical identification, whereas people of similar gender and age showed horizontal identification. As the Marilyn Monroe study[19] confirmed, women who read news about Diana's death showed stronger identification than did men[20], because they had vertical and horizontal identification together.

Identifying celebrities online or through social media is experienced by individuals in different ways[21], because digital media change the nature of relationships as they expand the scope of relationships between individuals and celebrities[22]. Since celebrities are passively represented through the mass media, whereas celebrities themselves actively express their own images through social media, the public and private images of celebrities exist simultaneously[23]. The mixed images of celebrities provide images of fancy stars and familiar friends together to their fans. In a digital media environment, a fan comes to experience a celebrity both as a star and as a friend.

Fans in digital space may be able to process information by reading about a particular subject or watching videos regardless of their intentions, but sometimes will search for relevant content according to their own will, and share information repeatedly with members within the Internet community[6]. Through these communication activities of fans, a star sometimes becomes a friend who speaks directly to the fans[21] or sometimes an object of worship that is awe-inspiring to the fans[24]. The mixed images of a star experienced through digital media provide a disparate dimension of the identification to the fans. Based on the discussions so far on communication behavior and identification, the following research issues are presented.

RQ 1. How does a fan's individual communication behavior using digital media relate to the identification with a fan object?

RQ 1-1. How does a fan's personal information processing / information seeking / information sharing relate to the vertical identification with a fan object?

RQ 1-2. How does a fan's personal information processing / information seeking / information sharing relate to the horizontal identification with a fan object?

4. Identification & Social Participation

The social participation of fandom includes pro-social collective activities as well as community activities associated with fandom. Putnam divided the social capital of the community into three types: social activities related to the interests of the community, donation and charity activities based on altruism, and civil movements oriented toward public interest, human rights, and social justice[25]. Based on these three types of community collective behavior, we attempt to explain the social participation of fandom from the perspective of spontaneity, persistence, and offline and online performance.

First of all, social participation can be classified as passive or active. The types of community participation can be subdivided into formal participation and practical participation depending on the degree of participation [26]. Second, One-time or steady participation in social action serves as a yardstick for measuring the outcome of participation. It can be divided into one-off participation, such as year-end donations and special events, and continuous participation, such as a fan club or sponsor activities[27]. The collective action of fandom can be categorized as participation in online virtual space and actual participation offline[2]. In the real world, participation includes various forms of collective action, ranging from stage performances, events, and gatherings of local assemblies to protest rallies and street demonstrations, which mean physical collective action. Spatial constraints and time limits are not a big problem in virtual space. Fans can click 'like' through online forums or bulletin boards, participate in online voting on fandom-related issues, or post messages to express their opinions collectively.

According to studies of celebrities such as Princess Diana[20], Magic Johnson[28], and Lady Gaga[29], fans perform social actions advocated by celebrities when they feel emotionally connected to them. Fans of singer Tory Amos, who had revealed her frank feminist beliefs through music, volunteer work through their identification with her[30]. On the contrary, Couldry & Markham argued that thinking popular culture provides an opportunity to entice the audience into the community culture of society is groundless, suggesting that groups that identify with celebrities have the lowest social and political participation[31]. It is still questionable that a fan's identification with a fan object is a determinant of social participation.

RQ 2. How is the identification with a fan object related to social participation?

RQ 2-1. How does identification with fans relate to voluntary social behavior?

RQ 2-2. How does identification with fans relate to continuous social behavior?

RQ 2-3. How does identification with fans relate to online participation?

RQ 2-4. How does identification with fans relate to offline participation?

Brown and his colleagues, who described the process of psychological attachment with celebrities in the perspective of identification, emphasized the need to examine what dimensions of people's psychological involvement with celebrities are related to attempts to imitate their actions[20]. Based on the discussions so far about identification and social participation, we further suggest, as a final research question, figuring out how the fan's social participation differs around the two dimensions of the identification in the digital media environment.

RQ 3. How does the relationship between identification and social participation differ in vertical identification and horizontal identification?

Ⅲ. Methods

1. Sampling & Procedure

This study conducted the survey on October 26, 2019, with fans who gathered for a BTS concert, "LOVE YOURSELF: SPEAK YOURSELF [THE FINAL]" at the Olympic Main Stadium in Seoul. According to ticket reservation statistics, the audience for the concert was composed of teenagers (33.7%), 20s (27.8%), 30s (23.1%), 40s (12.9%), and 50s (2.1%). The survey was completed on Google Survey. Fifteen helpers on the spot asked fans to access the online survey via a QR code to respond. The helper did not submit an online survey if the survey response time was less than five minutes. The total number of participants in the survey was 259, seven of whom were not ARMY, so the analysis was conducted with 252. The 252 ARMY members were mainly composed of women (94.4%), consisting of teenagers (41.3%), 20s (36.9%), 30s (14%), 40s (7%), and 50s (2%), which was almost identical to audience composition.

2. Survey Questionnaires

Respondents answered questions about whether they were ARMY and then demographic information and ARMY activities. Because the fan's sense of belonging to the fandom community is important as a motivational factor for identification[32], we used the scale normally used in fandom studies[33] to measure the degree of belonging to the fandom (3 items, 5-point Likert scale; e.g., The fact that I am ARMY is important to me). After confirming the hours of online and social media use for fan activities, respondents answered questions about their individual communication behaviors (information processing, information seeking, information sharing), the identification (horizontal, vertical), and social participation (voluntary, continuous, online, and offline).

3. Operational Definition & Measurement

3.1 Communication Behavior

We modified the measurement questions of information processing and information seeking from the study of Grunig[7] to suit the fandom activities using ARMY's digital media. Information processing was measured by 3 items (e.g., I tend to read articles or information about BTS online or through social media). The information seeking was also measured by 3 items (e.g., I tend to visit certain sites or social networking sites to find information about BTS). Because information sharing means that members share information through interactions within the fandom community[9], 2 items measured Fiske's productivity concept by applying it to ARMY activities in the digital media environments (e.g., , I post information about BTS online or through social media) and 2 items measured offline information sharing (e.g., I talk about BTS when I meet people around me). The questions about communication behavior all consisted of 7-points (1-point: not at all, 7-points: very much so).

3.2 Identification

We modified the measurement tools used in the study for celebrities[34], media characteristics[11], and an organizational-identification[35] to suit the situation experienced by ARMY with BTS (7-point Likert scale). Horizontal identification was measured in four aspects: (1) attitude or judgment aspects (e.g., I always think of BTS as a close friend); (2) cognitive concordance (e.g., I think I understand the meaning of BTS' song lyrics well); (3) emotional concordance and immersion (e.g., looking at BTS makes me feel like I see myself ); (4) sharing of goals and values (e.g., BTS' success is my success). On the other hand, vertical identification was measured like this: (1) attitude or judgment aspects (e.g., I think BTS is my ideal type); (2) cognitive concordance (e.g., I learn a lot from BTS songs); (3) emotional concordance and immersion (e.g., looking at BTS makes me feel like I see myself in the future); (4) sharing of goals and values (e.g., I want to follow BTS' way of life).

3.3 Collective Participation

The social collective action of fandom was measured by the degree of participation in terms of the degree of spontaneity, persistence, and online and offline performance, respectively measuring social activities such as ARMY gatherings, altruistic social activities such as charitable activities, and civic movements such as public campaigns. All items were measured on a 7-point Likert scale.

4. Data Analysis

We analyzed the collected data using the SPSS 21.0 program. We first checked the reliability and validity of the measurement items, verified the control effects through Hierarchical Regression, and tested the research questions. We used Linear Regression to analyze the relationship between ARMY's communication behavior and identification (RQ1), the association between identification and ARMY's social participation (RQ2), and the difference between horizontal and vertical identification on ARMY's social participation (RQ3). In addition, we used the Regression Analysis presented by Baron & Kenny[36] to test the mediating effect of the horizontal identification. The statistical significance of the mediating effect was verified by the Sobel test.

Ⅳ. Results

1. Measurement of Principal Variances

Inter-question reliability (Cronbach's α) for the measurement of variables, such as communication behavior, identification, and social participation, was very high, showing more than 0.8 in all items except for information processing (0.75) and information pursuit (0.79). The internal validity of all items has proven reliable.

2. Evaluation for Validity of Variances

If the variable's Average Variance Extended (AVE) value was 0.5 or higher, we judged that the variable satisfies convergent validity[37]. All of the variables in this study had convergent validity. In addition, the discriminant validity between the variables is satisfied if the AVE value is greater than the square of the correlation coefficient between the corresponding variable and the other variables[38]. The variables in this study also had discriminant validity.

3. Measurement for Controlled Variances

We examined the control effects by means of Hierarchical Regression when the independent variable of communication behavior affected the dependent variable of identification by injecting controllable variables, such as demographic and sociological factors (ARMY's gender, age, education level, average monthly cost for fan activities, when fan activity started, media usage time). We found that the explanatory power of the control variable (R² = .140) was increased (△R² = .283) when the control variable was injected with the independent variable (R² = .423). Thus, when communication behavior affected identification, we confirmed that control variables were controlled. The second Hierarchical Regression examined the control effects of control variables when the identification affected social participation. When a control variable was put in with an independent variable (R² = .429), its explanatory power was increased (△R² = .197) from that before the independent variable was put in (R² = .232). Therefore, it was possible to confirm that control variables were controlled.

4. Study Results

To verify RQ1, we used a Multiple Linear Regression[Table 1]. The two regressive models were found to be statistically significant (F = 44.73, p = .000; F = 41.54, p = 000), and there was no self-correlation (Durbin–Watson = 2.01, 1.97, respectively). The effect of all independent variables on vertical identification were statistically significant. On the other hand, a model with horizontal identification as a dependent variable could show that the effects of information processing activities (β = .26, t = 3.31, p = .001) and information sharing activities (β = .35, t = 4.98, p = .000) on horizontal identification were statistically significant, but the effect of information seeking activities (β = .05, t = 0.544, p = .587) on horizontal identification was rejected (p > .05).

Table 1. Multi Linear Regression result for RQ1

In sum, RQ1-1 was proved to show that information processing, information pursuit, and information sharing activities have a positive effect on vertical identification, whereas RQ1-2 showedthat only two types of communication activities, i.e.information processing and information sharing, had a positive effect on horizontal identification, whereas information pursuit activities did not.

Therefore, RQ1 has been partly verified as showing that digital media use has a positive effect on the identification with fan objects. In order to verify RQ2, we used Regression[Table 2]. It can be seen that the regressive model was appropriate and did not have self-correlation (F = 127.44, Durbin-Watson = 1.91). The regressive model, which identified the positive effects of vertical and horizontal identification on four types of social participation, also showed that all F values
and the standardized coefficient (β) of the independent variables were statistically significant. Therefore, it was statistically established that the identification with a fan object has a positive effect on social participation. Specifically, ARMY's identification with BTS positively affected voluntary, continuous participation, as well as online and offline participation. Thus RQ2 was supported.

Table 2. Regression result for RQ2

RQ1 and RQ2 confirmed that the identification mediated fans' use of digital media and social participation. To verify the mediating effect of identification, we used a three-step Regression Analysis. ARMY's communication behavior had a valid effect on the identification, which met the first condition (β = .62, p < 0.001). Subsequently, ARMY's communication behavior showed a significant influence on social participation, satisfying the second condition (β = .61, p < 0.001). Third, both the communication behavior and identification injected into the regressive model simultaneously showed a valid influence on social participation, meeting the third condition (β = .41, p < 0.001; β = .33, p < 0.001). Finally, the influence of communication behavior in the second stage (β = .61) was higher than that of the third stage (β = 41, β = .33), satisfying all the conditions of the mediating effect[Table 3].

Table 3. A three-step Regression Analysis result

We measured the significance verification of the mediating effects by the Sobel test to confirm that the mediating effect of identification was significant[Table 4].

Table 4. Sobel Test result

Therefore, the identification of ARMY with BTS between ARMY's communication behavior and social participation can be seen as having a mediating role.

In order to verify RQ3, which asks how the relationship between identification and social participation differs between the vertical one and horizontal one, we did a Multiple Linear Regression with horizontal and vertical identification as independent variables and social participation as a dependent variable [Table 5].

Table 5. Multiple Linear Regression for RQ3

In order to examine the influence of the two types of identification on social participation, the unstandardized coefficient (B) of the two independent variables, horizontal and vertical identification, was 0.62 and 0.24, respectively. This can be interpreted as an increase in social participation by '0.62' units when horizontal identification increases by one unit, whereas social participation increases by '0.24' when vertical identification increases by one unit. In addition, the standardization factor (β) again confirms that horizontal identification is a more important variable in predicting social participation than is the vertical one, because the horizontal one is 0.48 and the vertical one is 0.15 (p < .05). Thus the analysis of RQ3 showed that the effect of horizontal identification with the fan object experienced by the fan individual is greater than that of the social participation of vertical identification.

Ⅴ. Conclusions

We confirmed that the use of digital media leads to social participation, clarifying the process in which private feelings of individuals move toward public actions. Above all, a new approach to the identification has clarified the progress, which has not been shown in previous media audience studies.

When ARMY experienced vertical identification with BTS as a star, all types of communication actions positively affected the identification. Interestingly, if ARMY experiences horizontal identification with BTS as a friend, passive acts of processing and sharing of information have a positive effect on horizontal identification, but the active act of pursuing information by itself did not affected horizontal identification.

In the digital world, passive information sharing is synonymous with active information pursuit[6]. Individual passive communication actions that simply read or view content may be more effective than individual active communication actions, which require cognitive processes to understand or evaluate the meaning of the content[5]. In the midst of an infinite flood of content, individuals choose efficient methods that do not find information in person. The communication patterns that develop into a process of accidentally accessing information through the media (information processing), directly searching for information (information pursuit), and exchanging relevant information with others (information sharing) are no longer meaningless.

We also verified that ARMY's identification with BTS mediated the effect of communication actions on social participation. The ARMY’s identification with BTS as a star or friend was linked to participation in collective action, such as social gatherings, which expanded to social activities regarding human rights and environmental movements. The mediating role of the identification, which was theoretically presented in this study, has been confirmed.

What is noteworthy here is that horizontal identification with BTS induces more participation by ARMY in society than does the vertical one. This result indicates that the fan's perception of a fan object plays a crucial role in actual practice. ARMY actively participates in society, recognizing BTS as being similar to itself rather than following and admiring them. Horizontal identification based on private relationships with BTS is a much more important factor in ARMY's transformation into a pro-social public figure than is vertical one.

This study has revealed the process of personal feelings about a particular object experienced through the digital media into public action by mobilizing the existing fandom theory, media audience research, digital media computer-mediated communication research, and PR studies. It is expected that this approach, which proposed identification as horizontal and vertical, could help with research into collective action of people sharing a common concern in the digital environment.