Hashtags and Digital Movement of Opinion Mobilization: A Social Network Analysis/SNA Study on #BubarkanKPAI vs #KamiBersamaKPAI Hashtags

Abstract


INTRODUCTION
There is a recent digital activism phenomenon using hashtag (#) in postings on social media.Social media users show their attitude (agree or disagree) to an issue or policy by posting a comment along with a specific hashtag.This attitude generally emerges as a reaction to an event.For example, when public policy is announced and social media users disagree with the policy, their disapproval is expressed through anger by posting on social media.This hashtag created a community, where social media users who do not know each other nor follow each other, can discuss the same topic, even with the same attitude, in the virtual world (Bruns & Burgess, 2012).
These forms of digital activism are relatively new compared to the previously known online petition (e-petition) or digital social movement.The difference lies in the nature, form and characteristics of the digital activism.Digital social movement or online petition is generally led by a particular actor or social organization.This actor designs a campaign and later invites social media users (netizens) to get involved in the social movement campaign (Lindner & Riehm, 2009).Although it is carried out digitally, this campaign has a clear purpose and is led by an actor spearheading the movement.Meanwhile, this new model of digital activity is relatively more spontaneous.Social media users respond spontaneously to everyday events by writing comments on social media.Their activities are not controlled or led by a social movement actor (account), so it is more natural.Hashtag is used as a bridge which connects users to other users who have same interests and alignments.Barisione and colleagues (2017) referred to this digital activity as the Digital Movement of Opinion (DMO).DMO was born as a form of technological development, especially social media, where this media creates a virtual network between one user and another user.DMO is defined as an activity carried out by social media users by spontaneously commenting on emerging issues.Because it is spontaneous, this activity is not returned by social movement actors.This spontaneous nature also causes a short age of such digital activity, e.g.only a few days or even hours.Users are interested in an issue, and they will move on to another issue on the next day and so on.The most common forms of DMO are, for example, to make comments, make memes, send replies to a post and many more.Essentially, in the DMO, an issue is discussed spontaneously instead of being with specific goals or objectives as in the general social movement.
Digital activities of DMO also occur in Indonesia.Almost every day social media users comment on current issues ---ranging from corruption cases, forest fires, floods to traffic jams.Emerging events are commented on spontaneously and comments from users are replied to by other users.Therefore, the issue becomes a conversation on social media.The hashtags occupy an important position in this conversation, because these hashtags function like a "discussion topic".With hashtags, users can discuss and share statements with other users.
One case of the digital DMO activity is a controversy regarding the termination of children badminton audition by PB Djarum.After decades of granting badminton coaching scholarships, PB Djarum stopped this activity starting from 2020.Djarum pointed out that this decision was made because they did not agree with the KPAI (Indonesian Child Protection Commission) which asked Djarum to remove the Djarum brand in audition activities.The case was fiercely discussed by Twitter users in mid-September 2019.According to data from Politicawave, the total number of conversations about the case between 7 th and 11 th September is 23,728 posts.Interestingly, social media users are divided in their opinions, i.e. users supporting KPAI and users supporting PB Djarum.This support is expressed through the use of different hashtags, namely #Bubar-kanKPAI for those who support PB Djarum, and hashtag #KamiBersamaKPAI for those who support KPAI.For 5 days (7 th -11 th September), the hashtag #BubarkanKPAI is superior to the hashtag #KamiBersamaKPAI (see Figure ).The number of the hashtag #BubarkanKPAI is ten times more than KamiBersamaKPAI.This research wants to examine the role of hashtag (#) in the mobilization of digital opinion support, by studying the controversial case of children badminton audition suspension by PB Djarum.This case illustrates how different hashtags (#) can create different mobilizations.As shown in the picture above, the hashtag #BubarkanKPAI encourages more user mobilization to respond or have an opinion which is observed from the higher number of posts than #Ka-miBersamaKPAI.
Studies on hashtags and social movements have been conducted by a number of researchers.These studies looked at how hashtags played a role in social movements in the digital realm.For examples are studies on the use of hashtag #Austerity in Europe (Barisione & Ceron, 2017), hashtag for mobilization against #DilmaRoussef in Brazil (Calvo, Dunford, & Lund, 2016), hashtag #RefugeesWelcome (Barisione, et.al., 2017), hashtag #Ferguson (Bonilla & Rosa, 2015) and hashtag #BlackLives-Matter (Yang, 2016).The similarity of these studies lies on the observations to see how the hashtags create the mobilization of social media users and how hashtags encourage users to continue to be concerned with and involved with the issue.However, these studies do not answer the important question, i.e. can different hashtags (#) create different mobilizations?When an opinion movement uses a specific hashtag, will the mobilization of social media users be different if the movement uses another hashtag?This weakness arises because the studies generally use one hashtag only for their analysis.
This research attempts to address gaps or weaknesses from previous studies.The researcher used and compared two different hashtags, namely #BubarkanKPAI and #KamiBersamaKKAI.In order to see whether different hashtags have an impact on different mobilizations, the mobilizations among the two hashtags were compared.

LITERATURE REVIEW Digital Activism
The emergence of digital media (especially social media) has made interest in studies on social movements in the digital world increased sharply (Kaun & Uldam, 2016).Topics about digital social movements are given different labels, such as digital social movements or digital activism.Despite of making different labels, the studies generally focused on activities which were carried out by citizens in the social world to change things (issues, policies) using digital media (Kaun & Uldam, 2016).
Various studies on social movements in the digital realm have different focuses.In general, the focuses of the studies can be divided into two broad categories.First is studies which look at the digital world (online) as an extended arm of the offline social movements.These studies generally see the digital world as a medium used by the social movement actors to spread ideas and gain support from the public.Examples of these studies include researches by Bennett and Segerberg (2012), Morozov (2011), Loader, Steel, & Burgum (2015), and Anduiza, Cristancho, & Sabucedo (2014).A characteristic of these studies is a point of view that the digital activism is an inseparable part of the offline activities.The social movement actors see the digital world as a medium to create mobilization and support for any social movements (Van Laer & Van Aelst, 2010).
The second group is studies which see the social movements or digital activism as an independent and autonomous part of any offline social movements.These studies see that digital activism has its own traits and characteristics, and does not have to be related to the digital world.A real social presence in the social world (such as demonstrations, boycotts to protest, marches and so on) is not a prerequisite of the digital activism.In other words, the success of the digital activism is not measured by whether the digital activism has influence in shaping the offline movement.Examples of this type of study are studies by Dahlgren (2013) and Papacharissi (2015).

Digital Movement of Opinion (DMO)
The digital activism may take many forms.This diversity is shaped by technology.One of the digital activism forms is, for example, online petitions (Lindner & Riehm,200).In online petitions, leaders and members of any social movements use technology (a popular example is Change.org) to create support and mobilization.Other experts refer to the social movements using digital media with different terms, such as electronic advocacy and digital campaigns (Hick & McNutt 2002;Della Porta & Mosca 2005;Earl &Kimport, 2011).
Despite using different terms and labels, these studies see digital media as an inseparable part of the social movements.Studies generally see and link such digital media to the social movements, i.e. a movement initiated by a social movement actor (a person or public organization) and it has certain goals and interests.The studies also look at how digital media is used to create public engagement on any issues raised by the social movement actors.
Barisione and his colleagues developed an interesting concept of digital activism, which was named the Digital Movement of Opinion (DMO).This activism can be assumed to be different from the concept of using digital media for social movements (online petitions, electronic advocacy or digital campaigns).This concept mediates two important debates related to public participation (Barisione, et.al, 2017).First is the majority of public who generally express their opinions secretly (their opinions are generally known through public opinion surveys).Second, social movements are characterized by the activities of a small group of people with clear organizations, issues and leaders.Social media creates an activity which cannot be incorporated into such two forms of participation.When an issue appears, the social media users can spontaneously express their opinions (support or criticism) by posting them on social media.This post happens spontaneously, instead of being ruled or led by social movement leader actors.Through this post, the public's voice can also be heard clearly.
Barisione and his colleagues preferred the term opinion movement over social movements (Barisione & Ceron , 2017) .The term opinion is used because this form of activity is different from the concept of social movements in general.Social movements are generally characterized by activities which have specific objectives.To achieve their goals, social movements have a structure, create an issue, and an organization so the goals may be achieved.Social movements also have leaders who actively encourage participation from members or public to get involved in the movement.The social world users do differently ---they are not involved in the organization and not members of a particular social movement.Users spontaneously express their opinions in praising or criticizing certain policies.This digital opinion is very diverse and usually carried out by social media users daily, such as sending feedback, making a post, memes, commenting on the fellow users' opinion on an issue, and many more.
Barisione & Ceron (2017) identified 4 main features of the digital opinion movement.First are spontaneous and disorganized.Social media users spontaneously express their opinions and criticisms when reading news about an issue.No lead actor creates an issue and pushes the issue to the public attention.In the digital movement of opinion, users actively respond to issues by extending their opinions through posting on social media.Second, in terms of time, the age of this movement is not long.This characteristic is a consequence of the first trait.Because opinions are spontaneous and there are no organizing actors, attention to an issue can change quickly.Third, the opinion is generally homogeneous, i.e. black and white.Social media users express their opinions clearly, whether they support or criticize an issue or policy.Fourth, it is cross-sectoral because many groups or sectors are involved.
The concept of the Digital Movement of Opinion (DMO) was proposed by Barisione and colleagues because this form of digital activism was different from the commonly known social movements concept.The traditional concept on social movements, for example, was provided by Tarrow (1994), where social movements are characterized by activities carried out by a group of people with a collective goal, having solidarity and a clear identification of who the friends and foes are.Social movements also involve clear issues, organization, identity and goals to be achieved.This form of social movement is also found in the digital world, e.g. in the online petition movement.
The characteristic of this type of social movement is different from the opinion movement.If (traditional) social movements are characterized by clear collectivity and objectives, opinion movements are bound by attention to the same issue.Unlike some experts, Barisione does not see the activity of this opinion movement negatively as slacktivism or clicktivism (Bennett and Segerberg, 2012).The term is generally used to negatively describe forms of digital activism, such as users do not know the contentious issue, do not have a clear commitment and involvement with an issue, and many more.Barisione & Ceron (2017) stated that the activism of the opinion movement was a legitimate form which could be done by someone in the digital realm.

Hashtags and Mobilization
According to Yang (2016), hashtags played a role in digital activism.In fact, in the digital activism literature, a term known as "Hashtag Activism" emerged, namely a movement (in the form of approval or rejection) of something ascribed to it by using hashtags.One example of using hashtags in the digital activism is #Black-LivesMatter.This hashtag emerged as a form of protest from the social media users over the release of George Zimmer in the deadly shooting of an African-American teenager, Trayvon Martin (Yang, 2016).Users who protested against injustice mentioned the protest on social media using the hashtag.Another example of using hashtags in the digital activism is #Ferguson.This hashtag emerged as a form of protest over the shooting of Michael Brown by a police officer in Ferguson, Missouri.In the first week of Brown's death, millions of posts appeared on Twitter using the hashtag #Ferguson (Bonilla & Rosa, 2015).
Several studies found that hashtags had a role in creating mobilization.Berridge & Portwood-Stacer (2015) and Clark's (2016) studies, for example, show how gender equality and feminism activists use hashtags to create attention and encourage the involvement of social users to be active on the issues raised.
How can hashtags encourage mobilization in digital activism?Some experts give a different explanation.Bruns & Burgess (2012) explained the role of hashtags in creating imaginative communities.Social media users may not know each other (not following each other), but they seem to be in the same community and talking about the same things.Hashtag gives imagination about a large space where people can share opinions on the same topic.Through hashtags, a person can express approval or criticism of a topic without having to be a follower of certain social media accounts.
Meanwhile, Yang (2016) explained the importance of hashtag by using narrative theory.Successful social movements (including digital activism) need a story, because through this story the actor and the public feel involved with the issue being debated.Stories also make social movements as a daily problem faced by citizens.Through stories, the involvement of a person makes sense and is useful.According to Yang, even though hashtags contained short words or sentences, they had a narrative structure.Through this narrative structure, they seem to be in the same community, talking about the same people and events.Hashtags creates conflict, enemies and common heroes.Hashtags like #Lawan, or #Turunkan, for example, clearly identify who your friends are and who your foes are.

Hashtags and Digital Movement of Opinion
Hashtag has an important position in the digital movement of opinion (DMO).According to Barisione (et. al., 2017), hashtags functioned as anchors in the opinion movement.Hashtags can lure someone to express their opinions by posting on social media.A hashtag which is more encouraging for opinion is an emotional hashtag.This is consistent with the characteristics of the DMO, where hashtags can create spontaneous comments on an issue.This emotional hashtag is a form of anger or praise which can provoke compassion (pity) or anger from the social media users.According to Barisione (et. al., 2017), this emotional trait was both strength and weakness.Its power, through emotional hashtags, can encourage spontaneous opinion, so the discussion of an issue can be highly trending.Meanwhile, the discussion of an issue is short in which it becomes the weakness.The social media users will turn to other issues which have higher emotional value.
The hashtags, which are successful in provoking opinion, are emotional and in general, have clear frames (Barisione, et. al., 2017).Hashtag clearly identifies what causes of the problem are, who the causes of the problem are, and what the recommended solution of the problem is.Barisione (et. al., 2019), for example, gave an illustration of #RefugeesWelcome.This hashtag clearly defines the problem (discriminatory attitude), identifies the cause (right-wing politician or party) and recommends the solution of the problem (refugees accepted by the state).Clear frames make it easy for users to engage with the issue.Those who agree with an issue (as clearly explained in the hashtag) will also share their opinions on social media.

CASE
This paper focuses on the importance of hashtags in the digital movement of opinion.More specifically, this paper wants to test whether the use of certain hashtags has an impact on the mobilization of digital opinion.For example, an event or issue which has been given a hashtag filled with emotions will provoke more comments or opinions from the social media users than those which are ethical (not emotional).To answer this question, this study chose a case of discussion on social media related to PB Djarum's decision to stop badminton audition for children due to a conflict with the Indonesian Child Protection Commission (KPAI).This issue has an emotional element which stays, and provokes pros and cons among the social media users.The social media users express their attitude by including hashtags in a post on Twitter.Users who support PB Djarum use the hashtag #Bubar-kanKPAI.Meanwhile, those who support KPAI use the hashtag #KamiBersamaKPAI.
This case took lace in mid-September 2019.The case began with a polemic on children badminton audition between KPAI and PB Djarum.PB Djarum's badminton audition is a talent search for badminton.It has been held by the Bhakti Olahraga Djarum Foundation since 2006.In this audition, participants (children) wear shirts and uniforms written with Djarum brand.According to the KPAI, the PB Djarum's action is a violation of the Law of the Republic of Indonesia Number 35 of 2014 on Amendment to Law Number 23 of 2002 on Child Protection.This audition is a form of child exploitation by cigarette products.Through this audition, cigarette brands have been introduced since early age and the Djarum brand is associated with positive things.Thus, it is feared that children will consider cigarettes as a good product.Djarum itself rejected KPAI's opinion.According to Djarum, the audition organizer is not the Djarum cigarette company.Rather, it is the Djarum Foundation which is a different institution from PT. Djarum as a cigarette company.Djarum also stated that, from the audition to the quarantine (when someone is accepted as a Djarum athlete), the athlete is strictly forbidden to smoke.KPAI does not prohibit the auditions.What is questioned by KPAI is the use of the Djarum brand attached to the audition.According to KPAI, Djarum can still hold auditions by not using Djarum's brand name in the auditions.The polemic between Djarum and KPAI met a deadlock.The climax is PB Djarum officially announced the dismissal of public auditions start-ing from 2020.This announcement was made in a press conference on 7 th September 2019 at Aston Hotel, Purwokerto.They stopped the event until a polemic with the Indonesian Child Protection Commission (KPAI) found a solution.News regarding the termination of the audition was intensely discussed by the social media users (netizens).Those who disagreed with KPAI's action expressed their disapproval through criticism and verbal abuse by using the hashtag #Bubar-kanKPAI.This hashtag had topped the conversation (trending topic) for several days on the Twitter timeline.Meanwhile, other social media users, who assessed that the KPAI's statement was correct, defended and supported the KPAI's decisions.They expressed their support through posting on social media (especially Twitter) with the hashtag #KamiBersamaKPAI.
Those who differ in their posts on Twitter use different hashtags.The use of these different hashtags provides a good opportunity to test whether certain hashtags have more interest in social media users.

METHOD
To answer the question how hashtags affect on the mobilization of social media users, the SNA (Social Network Analysis) method is used.Researcher compared the network formed from two hashtags (Disband KPAI vs #KamiBamaKPAI).
Hashtag is said to have successfully mobilized social media users (netizens) if they have better network structure characteristics -characterized for example by density, reciprocity, diameter, and so on.
This method basically intends to describe the structure and relationship network of the actors (in this case, the social media users).This method looks at the relations among the actors (nodes/ social media accounts) in a particular social structure.This method will illustrate how the network structure of social media users (Twitter) is related to the polemic between KPAI and Djarum in the social media.The data analyzed in this study is nudges (tweets) on Twitter which use the hashtags #BubarkanKPAI and #KamiB-ersamaKPAI.The data were collected from 7 th -11 th September 2019.The total amount of data included in this study is 1,000 tweets for each hashtag.The research process was carried out in two steps as follows.First step is data crawling on Twitter which is related to both hashtags.Data were collected by using Netlytic software.The author entered the keywords, i.e. the hashtags #BubarkanKPAI and #KamiBersamaKPAI, on Netlytic.The data obtained were later cleaned.The total number of the data analyzed is 910 for the hashtag #Bubar-kanKPAI and 932 tweets for the hashtag #Ka-miBersamaKPAI.The shortcoming of the Netlytic software is its disability to take the whole discussions on Twitter.Therefore, the amount of data (tweets) analyzed in this study is only a fraction of all tweets which use both hashtags.
Second step is an analysis on the social media conversation data which had been obtained in the first stage.A network analysis is carried out in three levels.First is the network structure.Analysis at this level is intended to illustrate the shape and structure of the network.The network level does not talk about actors (nodes), but it talks about the network structure as a whole.Second is group level.This level describes the grouping in the network.How do the actors (nodes) in a network form one group, which is different from other groups?Third is the actor.This level identifies the position of the actor in the network and describes the dominant (central) actors in the network.The aspects analyzed for each level are as follows.

RESEARCH OUTCOMES
The writer will first describe the network analysis of the hashtags #BubarkanKPAI and #Ka-miBersamaKPAI.Later, it will be followed by a comparison of the two hashtags.

#BubarkanKPAI
The case began with a press conference held by PB Djarum on 7 th September 2019 at night.In this press conference, PB Djarum said that they would stop badminton audition for children starting from 2020.The media in Jakarta generally only reported on the case the following day, i.e. 8 th September.Interestingly, the conversation about this case has started to be trending on Twitter since 7 th September in the evening.On this date, there were 22 Twitter posts using the hashtag #BubarkanKPAI.The number of posts using this hashtag increased to 1,513 posts on 8 th September.The highlight of the post was 9 th September with 14,421 posts.The number of posts with this hashtag dropped to 4,894 on 10 th September and dropped again to 275 on 11 th September.In total for 5 days (7 th -11 th September), there were 21,125 Twitter posts using the hashtag #BubarkanKPAI.Following picture presents a communication network for the #BubarkanKPAI hashtag which was created using the Netlytic software.sible party for the suspension of auditions.They also asked the PB Djarum to remain holding the auditions and also demanded the government to dissolve the KPAI, which was deemed as the source of the problem.Twitter users have also posted stories about poor children who had to forget their dreams due to such suspension.The successful story of badminton athletes from PB Djarum in getting international achievements also coloured many posts by the social media users who used the hashtag #BubarkanKPAI.
Twitter posts with the hashtag #BubarkanK-PAI generally did not raise the KPAI's recommendation, where PB Djarum could remain holding the badminton audition by eliminating the brand in the Djarum audition.The posts actually blamed KPAI as the party who was looking for PB Djarum's mistakes and did not appreciate the sacrifice of PB Djarum who had spent big money to hold an audition.This frame can be seen from the keywords which are widely used by the social media users using the hashtag #Bubar-kanKPAI.Keywords used include: #Bubarkanlentera anak, anak (children), audisi (audition), generasi (generation), bangsa (nation), and many more.PB Djarum was described positively as seen from positive keywords.Meanwhile, KPAI and Lentera Anak (a public organization which is concerned with the issue of child exploitation through badminton auditions) were described negatively, through the use of negative words --such as disperse.The hashtag #BubarkanKPAI caught the attention of the social media users.This can be proven from the number of social users who expressed disagreement with the KPAI's attitude by using this hashtag.In addition to the number of posts, the success of this hashtag is also observed from the network structure as shown in the Table .The diameter is 5.This shows that the post spread from one user to another user (social media accounts) in 5 stages.The density and reciprocity is low, i.e. close to 0 (density = 0.003011, reciprocity = 0.005581).A low density indicates a low interaction among users (social media accounts).Posts are also not bidirectional, which is characterized by a low reciprocity.When users (social media accounts) expressed their disagreement with KPAI, they tended to be one-way as they did not respond to posts from other social media users.The interesting point is a relatively low centralization (0.288600) and relatively low modularity (0.477200).Low centralization shows that there is no dominant actor (social media accounts) who directs the contents of the conversation on Twitter users' social networks.In the meantime, a low modularity (leading to 0) shows the emergence of many groups (clusters) related to the conversation.Conversations are not dominated by dominant actors (accounts) because they are spread and form small clusters.The picture shows several clusters formed from the conversations on social media using the hashtag #BubarkanKPAI.From this picture, we can see that the large network of users who use hashtags is divided into small clusters.This type of network structure shows that a conversation with the hashtag is relatively scattered.The conversations are not commanded by several social media accounts.Instead, the conversations spread to multiple accounts and clusters.The table shows the actors (social media accounts) who are most often mentioned (mention, reply, retweet) in the conversations using the hashtag #BubarkanKPAI.PB Djarum (@ pbdjarum) and KPAI (@kpai_official) accounts are the most mentioned.In terms of personal accounts (netizens), the mentions relatively spread to several social media users.

#KamiBersamaKPAI
The hashtag #KamiBersamaKPAI appearred as a reaction to the hashtag #BubarkanKPAI.The hashtag #BubarkanKPAI was the most talked about topic (trending topics) on 8 th to 9 th September 2020.On 7 th and 8 th September when many social media users posted tweets using the hashtag #BubarkanKPAI, the hashtag #KamiB-ersamaKPAI has not yet appeared.This hashtag has only appeared on 9 th September.This hashtag seems to have been used to compensate for posts on the netizens' anger at KPAI which peaked on 9 th September.On this date, there were 2,257 posts using the hashtag #KamiBer-samaKPAI.Posts using this hashtag decreased to 332 on 10 th September, and dropped again to 14 posts on 11 th September.The total number of posts using the hashtag #KamiBersamaKPAI between 7 th to 11 th September is 2,603.Following image shows the visualization of the #KamiBer-samaKPAI hashtag network structure.
Compared to the hashtag #Bubarkan KPI, the hashtag #KamiBersamaKPAI has been less successful in mobilizing Twitter social media users.This can be observed from a low number of posts using this hashtag, i.e. 2,603 posts (compare to the number of posts with the hashtag #Bubar-kanKPI which reaches 21,125 posts).The number of posts with the hashtag #BubarkanKPI is tenfold more than the posts with the hashtag #Bubar-kanKPAI.The posts with the hashtag #KamiB-ersamaKPAI are intended as a counterweight to the posts with the hashtag #Bubarkan KPI, so it makes a low number of posts.
Users who use the hashtag #KamiBersamaKPAI     The hashtag #KamiBersamaKPAI is used by the social media users (netizens) to counter negative views about KPAI, and also confirm the dangers of cigarettes to children.If the use of hashtags is seen as a battle (#BubarkanKPAI versus #KamiB-ersamaKPAI), the hashtag #KamiBersamaKPAI is relatively losing.Apart from the fewer number of posts, it can also be seen from the following network structure (see Table ).For the density and reciprocity, similar to #BubarkanKPAI, the hashtag #Ka-miBersamaKPAI has a low density and reciprocity.Interaction between users is relatively small and one-way.The diameter of the hashtag #KamiBer-samaKPAI is only 3, which can be construed that this hashtag is relatively short, i.e.only 3 times (steps).Modularity is relatively high (0.713900) which can be interpreted that the homogeneity is quite strong.The conversations using this hashtag do not sufficiently spread to many users.The picture shows the clusters formed by using the hashtag #KamiBersamaKPAI.When they are compared to the hashtag #BubarkanKPAI, the clusters of the hashtag #KamiBersamaKPAI are more centered.This can be read that the conversations using this hashtag are more dominated by certain social media accounts, which means it less spreads as in the hashtag #BubarkanKPAI.Following table shows the dominant actors (social media accounts) in the conversation using the hashtag #KamiBersamaKPAI.Some social media accounts which are frequently referred to (retweet, mention or replay) are @k1ngpurwa, @ p3nj3l4j4h and @yusuf_dumdum.

Hashtags Comparison
The table presents a comparison between the hashtags #BubarkanKPAI and #KamiBer-samaKPAI.Observing the density and centralization, there is no difference between the two hashtags.Density refers to the density of relationships between social media accounts in a network.This measure shows the intensity among social media accounts in communicating (Golbeck, 2013).Both the hashtag #BubarkanKPAI and #KamiBersamaKPAI have a low density (density of #BubarkanKPAI = 0.003011; #Ka-miBersamaKPAI = 0.003120).Meanwhile, centralization refers to the concentration level of the network at certain actors (nodes).Regarding this aspect, both hashtags are the same (centralization of #BubarkanKPAI = 0.288600; #KamiBer-samaKPAI = 0.242100).
This similarity in the density and centralization shows that this case is a digital movement of opinion (DMO).In the DMO, there are no leaders of social movements.The social media users without being ordered by an actor or social movement leader deliver supports to and rejections against the case.The absence of any leaders or actors who give this command can be seen in the data where the density is low (indicating there is no interaction among social media accounts) and no social media accounts become the centre.Information about the case is obtained from many sources, instead of being centered on one account.
In addition to the similarities, the two hashtags also have differences, especially in terms of diameter, reciprocity, and modularity.Diamater is the farthest distance between one social media account and another account in a network (hashtag).This measure shows how far a message (post) spreads from one account to another (Golbeck, 2013).The diameter is 3 which can be construed that a post spreads a maximum of three steps, and so on.Diameter describes the extent of the message (post) distribution.From this size, the diameter of #BubarkanKPAI is higher compared to #KamiBersamaKPAI.If it is simplified, the distribution of messages within the hashtag #BubarkanKPAI network is wider.
The hashtag #BubarkanKPAI also has a higher reciprocity value (reciprocity #BubarkanKPAI = 0.005581; #KamiBersamaKPAI = 0.000000).Reciprocity is a measurement which describes the twoway relationships among social media accounts in the network.Reciprocity is calculated by looking at the proportion of accounts in a reciprocal conversation compared to the total number of conversations.The reciprocal value of the hashtag #BubarkanKPAI shows that accounts (netizens) which use this hashtag are relatively more twoway (reciprocating messages and posts) compared to the hashtag #KamiBersamaKPAI.
From the modularity aspect, the hashtag #BubarkanKPAI is relatively more homogeneous.The modularity size is related to grouping in a network.Modularity provides estimates whether a network consists of a group of accounts which form clusters (values close to 0) or overlapping accounts (values closer to 1).The modularity of the hashtag #BubarkanKPAI is lower (modularity of #BubarkanKPAI = 0.477200; #KamiBersamaK-PAI = 0.713900).This data illustrates that the accounts which use the hashtag #BubarkanKPAI are relatively more grouped.Data on the network structure above shows the hashtag #BubarkanKPAI is more successful than the hashtag #KamiBersamaKPAI.The superiority of the hashtag #BubarkanKPAI is marked by better reciprocity, modularity and diameter.The hashtag #BubarkanKPAI is more successful in mobilizing the opinions of social media users (netizens).Other evidences of the success of the hashtag #BubarkanKPAI are described as follows.First, there are relatively more dominant actors in the hashtag #BubarkanKPAI.Social media accounts which discuss the hashtag are more numerous and varied than the hashtag #Ka-miBersamaKPAI.Second, the hashtag life cycle is longer.The hashtag #KamiBersamaKPAI only lasted for 1 day.In the meantime, the hashtag #BubarkanKPAI lasted for almost 4 days.

DISCUSSION
The controversy over the termination of badminton auditions by PB Djarum shows the emergence of the digital movement of opinion (DMO).The characteristics of the DMO as described by Barisione and colleagues (2017) can be found.First, the conversation takes place spontaneously as a reaction from the social media users.The statement from PB Djarum during the press conference which announced the suspension of the badminton auditions was immediately responded to by the social media users.The users (social media accounts) posted tweets to show support and rejection to PB Djarum's decision.Posts are not regulated or controlled by certain social movement actors.Second, the age of the conversations is relatively short, which is 3-4 days.The peak of the conversation occurred on 9 th September (with a total of 16,678 posts).On the next day, the number of the posts has decreased and even disappeared after 11 th September.A relatively short life cycle of hashtags is an important feature of the DMO.As the conversation arises due to a spontaneous reaction instead of deep concern over the issue, the conversation will be replaced by another more interesting topic on the next day.This is different from traditional social movements, where social movement actors try to keep the issue as the public attention.In the case of this controversy, spontaneous issues are not part of certain social movements.
Another important feature of the DMO reflected in this case is homogeneous opinions.The social media users express their opinions on an issue homogeneously, i.e. in black and white.The social media users are divided between those supporting PB Djarum (using the hashtag #BubarkanKPAI) and agreeing with KPAI (using the hashtag #KamiBersamaKPAI).
This controvery of badminton audition suspension by PB Djarum is a thought-provoking issue to examine how the hashtags can play a role in mobilizing opinions among the social media users.Each social media user with his respective attitude expresses opinions using different hashtags.Although this opinion movement is not encouraged by any social movement actors, hashtags play a role in mobilizing the supportive opinion.Under certain hashtags, the social media users try to attract the attention of other users to get involved in expressing opinions.Hashtag serves as an anchor so the social media users have the same imaginary community, where each user who does not know each other nor follow each other (follower) can discuss with the same topic.
The results of this study show that #Bubar-kanKPAI is more successful in mobilizing opinions compared to the hashtag #KamiBersamaK-PAI.The success of the #BubarkanKPAI hashtag is marked by several aspects including more number of posts.The number of posts with the hashtag #BubarkanKPAI for 5 days (7 th -11 th September) is 21,125, which is ten times more than posts with the hashtag #KamiBersamaK-PAI (2,603 posts).Another aspect which shows the success of the #BubarkanKPAI hashtag is reciprocity, diameter and modularity.The hashtag #BubarkanKPAI reaches more social media users and is more widespread.
Why is the hashtag #BubarkanKPAI more successful in the mobilization than #KamiBersa-maKPAI?Using the framework proposed by Yang (2016), the hashtag #BubarkanKPAI provokes more emotions and creates narration.Emotions and narratives are very important in a DMO because opinions in a DMO are spontaneous and voluntary.Emotional hashtags can lure users to spontaneously react to an issue.While narration is important because it can bind the users as the narration provides conflict, it presents a common hero as well as a common enemy.With narration, the statement delivered by the users (including anger and verbal abuse) gets justified because the statement is based on strong reasons.
For 5 days (7 th -11 th September), the hashtag #BubarkanKPAI was filled with various narratives which made this hashtag persist in mobilizing opinions.An example of such narratives is a narrative on children from poor families who rise to the fame (become successful) after they received a scholarship from PB Djarum.Another narrative which appeared in the social media conversations is about costs which must be incurred by PB Djarum in developing and nurturing outstanding athletes.Do not forget the emergence of the nationalism narrative, such as how PB Djarum has raised Indonesia's good reputation through the achievements and triumphs of her athletes in the international world.A rich narrative which binds the social media users like this does not appear in the hashtag #KamiBersa-maKPAI.The hashtag #KamiBersamaKPAI has not been successful in creating a narrative which makes the users feel bound and involved with the issue.
The success of the #BubarkanKPAI hashtag can also be explained through the Barisione's framework (et. al., 2019) regarding the strength of the frame in a hashtag.Successful hashtags are not only emotionally capable of forming spontaneous responses from the social media users, but they also have a clear frame.This can be related to the characteristics of the DMO, where in the DMO the conversation about an issue tends to be black and white.If social media users do not agree (support), they will choose the other attitude (against).This kind of black and white attitude will be more easily mobilized by hashtags which have clear frames, i.e. a frame which clearly shows where the problem is, who the cause of the problem is and how to resolve the problem.
The hashtag #BubarkanKPAI clearly puts the controversial issue of badminton auditions as an issue of KPAI.KPAI is the cause of the problem.This hashtag also clearly and expressly shows the attitude on the solution of the problem (by using the word "bubarkan" [disperse]).In other words, even though it consists of words or phrases, the hashtag #BubarkanKPAI clearly shows attitude and how the problem is seen, including problem solving.Such clear frames do not appear in the hashtag #KamiBersamaK-PAI.This hashtag does not clearly indicate what the problem is and who the party causing the problem is.Recommendations for solving the problem (using the word "kami bersama" [we together with]) are not accompanied by strong reasons why KPAI must be supported.The hashtag #KamiBersamaKPAI is more of a response, and therefore does not show a clear frame.The absence of a clear frame does not encourage the mobilization of users who agree with KPAI's reasons.
In order for the measures taken by KPAI obtaining support (mobilization) from the social media users, other hashtag can actually be used where such hashtage should be more emotional and able to create a clear frame, e.g.#Hentikan-EksploitasiAnak. If this hashtag is used, it will likely be supported, because it has an emotional dimension and also a clear frame.Through this hashtag, the case is seen in the framework of child exploitation, where PB Djarum is a form of child exploitation.This hashtag clearly defines the problem (child exploitation) and put PB Djarum as the cause of the problem.Then, the KPAI's statement which demands for the removal of the Djarum brand attributes will then be seen as a child exploitation problem solution.Unfortunately, this hashtag is not used.The hashtag used by the social media users who defend KPAI is #KamiBersamaKPAI which turned out to be inattractive for the social media users in order to engage and participate in providing their opinions.
The victory of the hashtag #BubarkanK-PAI has serious implications for the public, because information is dominated by the PB Djarum's statements.The public (at least social media users) do not get complete information, especially how to see this controversy from the perspective of the dangers of smoking and children protection.The framework of the dangers of smoking and how cigarette brands use auditions to strengthen brands do not emerge in the social media talks.Child protection also does not appear in the social media conversations.Although PB Djarum provides scholarships for children, this activity must be viewed critically as a form of child exploitation, i.e. how children are used to strengthen the cigarette company's brand as a company which is concerned with the development and health of children.Such perspectives do not emerge much in the social media conversations, because the social media space is more filled with those who support PB Djarum with the hashtag #BubarkanKPAI.In other words, if the hashtag is seen as a battle, the victory of the hashtag #BubarkanKPAI can also be interpreted further as a victory of the frame which tends to take side on the cigarette companies rather than the child protection.

CONCLUSION
This research shows the importance of hashtags in the digital movement of opinion.The hashtag #BubarkanKPAI is able to create a mobilization compared to the hashtag #KamiBer-samaKPAI.The mobilization in this study was measured by the number of posts and the network structure formed.The success of the hashtag #BubarkanKPAI lies on its more emotional trait, its ability to create narrative imagination, and its clear frame.The results of this study have implications on how actors use hashtags to get more support in the digital realm.The social movement actors must pay more attention to the use of hashtags, so the issue can be more supported by social media users.

Figure 1 .
Figure 1.Total Number of Conversations

Figure
Figure 2. Research process

Figure 5 .
Figure 5. Clusters in the #BubarkanKPAI Network try to respond to the hashtag #Bubarkan KPI with a new frame.The contents of their posts defend KPAI.According to the contents of the post, this case would actually not be protracted if PB Djarum was willing to not use the Djarum brand in the audition program ----for example, erasing Djarum's writing on the badminton shirt or court.KPAI does not prohibit the badminton audition.Rather, it con-

Figure 9 .
Figure 9.Comparison on Dominant Actors and Hashtags Life Cycle

Table 3 .
Dominant Actors in the #BubarkanKPAI Network

Table 6 .
Comparison of the Hashtags Network Structure