Four Strategies of Social Media Use Among Indonesian Politicians

. This study aims at unveiling strategies based on the patterned use of social media by politicians. Using an interpretive case study involving Indonesian politicians from national, provincial, and district level parliaments, the study identifies four strategies: nominal, instrumental, manipulative, and genuine. The selected strategy is reflected by internal and external affordances of social media perceived by the politicians, and influenced by a variety of constraints. These include poor Internet connection, limited capabilities of politicians, low ICT literacy among constituents, security issues, personal attack, unsupportive regulation, and fake accounts.


Introduction
In the last 15 years, social media as user-generated platforms have become rapidly gaining popularity on the Internet [1]. Social media manifest in various forms such as blogs, social networking sites, microblogging site, video sharing site and many other user-driven platforms [2]. As social media communication for eParticipation gets denser, more complex and more participatory, citizens are getting more access to information and even more opportunities to engage in political discourse [3]. eParticipation involves the use of information and communication technologies (ICT) to deepen political participation by enabling citizens to connect with one another and with their elected officials [4] as well as to better involve citizen in a decision-making process by the government [5,6]. The use of social media in eParticipation may provide various "action possibilities" (i.e. affordances) that are perceived differently by users since one object can produce different outcomes [7]. Actualized affordances explain action taken by the actors as they take advantage of the use of ICT to achieve goals [8]. For instance, for politicians, social media may provide various affordances, such as value sharing, opinion gathering, political networking, campaign funding, vote gathering, and promoting participation [9,10]. From the constituent perspective, it may provide a means for political participation. Furthermore, previous studies on social media applied affordance concept to examine implication and relationship between these new technologies within organization [7,11,12], government and public bodies [13] and also political parties [10,14].
The vast majority of studies on this topic focused on the functional affordances at the organizational level (such as government agencies), which are tightly coupled with features and functionalities of social media, and only few so far has been scrutinizing affordance for practice which "sees affordances as both dispositional and relational and explore the affordances of social media as embedded in and emergent from social processes within organizational boundaries" [ 15 p.307]. Hence, there is a need to further explore social media affordances emerging from practice actualized by individuals (e.g. politicians). In this study, we attempt to fulfil this void by studying the affordances of social media perceived by politicians and unveiling the possible strategies they chose.
More specifically, we seek to answer: What are strategies adopted by politicians in using social media for eParticipation initiative? In doing so, we conducted an interpretive case study involving Indonesian parliament members from three levels of centrality: national, provincial, and district levels. We have identified both internal and external affordances of social media use [9]. However, we have not yet further analysed possible patterns that may emerge from the identified affordances, which lead to specified strategies. Further, to better understand and conceptualize the identified strategies, we borrow the concepts of eParticipation (i.e. information, consultation, and active participation) [16] to guide the discussion.
This study is important for several reasons. First, conceptually, it contributes to the rather neglected study of social media affordances in the political arena at the individual levels. Second, contextually, it provides a picture of social media use in eParticipation from less articulated context, which is Indonesia, reported in literature. Third, practically, the findings may be useful as references in formulating suitable strategies for a specific context.

Theoretical Premises
The increasing interest of social media adoption for eParticipation makes it relevant to look into what these new media afford in the context of encouraging more citizen participation in decision-making processes. The notion of affordance was first coined by Gibson [17] who explained that affordance emerged from the interaction of an actor related to the surrounding of the actor (the environment). Such view implies that multiple affordances of the same object may surface relating to different perspectives of the actor [8,14,17]. Information systems scholar argued that affordances exist as a relationship between an actor and an artefact reflecting possible actions on the artefact, dependent to the capabilities of the actor [18]. However affordances may also constraint an actor to perform an action. Indeed, affordances emerged from object features and dependent on the actor's capabilities, in result these potentials serves the dual concept of affordance, it can be enabling and constraining at the same time [18].
The constraints serve as the reference for both feedbacks from behavior and feedforward to anticipate future behavior. The fourth dimension is the value constraints, which are necessary to differentiate the possibilities for behavior in terms of "good or ill" [19].
With the changing dynamic of affordance perspectives in the information systems field, scholars offered an alternative standpoint of affordance-for-practice where the analytical focus shifted from technology to practices [20]. Here affordance is always emerging from technology-involved practices associated to the experience, skills, cultural understanding and capabilities of the actors, which is most relevant to examine specific users with certain needs, goals, and practice in particular social, cultural, and historical context [15,20,21]. Previous study showed affordances approach was able to display multiple benefits of social media use by political parties in which revolutionizing political campaign by affording many different issues and interest to appear in a political discourse [14].
Our previous study identified affordance for practice by politicians, which includes internal and external social media affordances. Internal affordances refer to affordances arising from practices, which involve social media that are perceived between politicians at the parliament. External affordances are perceived capabilities of social media between politicians and constituents. We will use the concept of internal and external affordances to see whether it may or may not influence the chosen strategy of social media use in eParticipation among politicians.
In doing so, we want to see whether politicians in affording social media in practice are also motivated by perceived constraints of social media. By analysing the relationship between identified affordance for practice and perceived constraints from the perspective of politicians and citizens, we hope that it will help us in conceptualizing the identified strategies of social media use.

Research Context
This empirical study is performed in Indonesia aiming at identifying different strategies of social media use by Indonesian politicians. Indonesia is a country where Internet users are reaching 100 million in 2016, almost half of the total population of 250 million. Different level of parliament in three distinct areas includes national parliament in Jakarta (DPR RI), provincial parliament in Special Region of Yogyakarta (DPRD DIY) and district parliament in Gunungkidul (DPRD Gunungkidul) may give different perspectives in promoting participation via social media by politicians. The parliament at any level is obliged to perform the function of legislation, budgeting, and supervision where the representatives shall give priority to the interest of the constituents. 1 Various insights from different geographical areas in Indonesia may enrich our understanding of the politicians' strategy in harnessing social media for ePartici-4 pation influenced by several different contextual issues such as ICT infrastructure, constituents' characteristic, customary, and cultural background in the electoral base.
In addition, having considered the history of democracy in Indonesia, which experienced a government transformation from authoritarian to democratic and open system in 1998; it is interesting to look into the development of citizen participation after the ICT, e.g. social media, is entering the scene.

Data collection
By adopting interpretive case study approach [22], we conducted in-depth interviews, from January to September 2016, with key actors of eParticipation in three levels of parliament in Indonesia: five politicians in the national, five in the provincial, and four in district parliament. During the interviews, we emphasized on how politicians perceived social media affordance in relation to their choice of strategies for eParticipation. In addition, the data were also acquired from interviewing three political activists and five political strategists with national and local experiences. In this case, political activists represent citizens' perspective since they came from citizens' element of eParticipation whereas political strategists represent politicians' perspective as they usually are hired by politicians. In total we interviewed 22 informants. We expected that the variety of perspective brought in by the informants provided a more complete picture about the topic under investigation. We also include secondary data from social media such as status updates, blogs, and comments related to eParticipation in our analysis to enrich our understanding of the phenomena.

Data analysis
We used the concept of internal and external social media affordances obtained from our previous work [9] and the concept of eParticipation includes information, consultation, and active participation [16] to guide the data analysis. We revisited the identified internal and external affordances from the collected data. We coded the transcription in an iterative manner where we go back and forth to code the actualized affordances with the matched internal and external affordance concept. In addition to the coding process, partly we used NVivo to help us identifying important concepts and categories. We used hermeneutic circle as a sense-making strategy in which we moved back and forth between detail and sense of a whole, brought together two different realms, textual and social, in the process of interpretation [23]. After the identification of internal and external affordances, we looked for possible strategies. In this study, inspired by the work of Mintzberg [24], we define strategy as "plan, ploy, pattern, position, or perspective adopted by politicians in using social media to achieve specific goals". In short, a strategy may be developed in advance and with purpose (plan), serve as a means to win the competition (ploy), repetition of successful initiatives from the past (pattern), a means of locating actions in the environment (position), and actions influenced by a worldview (perspective).

4 Findings and Analysis
Our previous data analysis found a set of affordances offered by social media used by politicians [9]. When further analysing the findings, we revisited the data. Affordances emerge from the coupling between politicians and social media can be classified into two groups: internal and external affordances. The internal affordances of social media use among politicians include idea storming, information storing, information sharing, peer entertaining, and sending event invitation. The external affordance of social media when used in connecting politicians and their constituencies, include value sharing, opinion gathering, political networking, personal branding, maintaining political existence, and promoting participation. When re-analysing the data, we found additional external affordance of mobilizing supporters, which we did not reveal in the previous study. As can be seen in Table 1, while internal affordances are mutually exclusive among the identified strategies, perceived external affordances do not exclusively fall under one strategy category only (except for nominal).

Strategies of social media use
After the identification of internal and external affordances, we looked for patterns. We grouped politicians reported a set of similar affordances or were being reported by their social media strategist about their social media. One politician may belong to more than one group. In this case, we grouped them into only one group that best represents strategy they adopted. This technique resulted in four categories of strategies: nominal, instrumental, manipulative, and genuine (see Table 1). The strategist also identified that some politicians obviously adopted genuine strategy which are not many.

Understanding the chosen strategies
The question raises is then, how to understand why certain strategy is chosen by politicians? The relationship between the perceived affordances and the strategies are explained using other findings from the field, especially those related to contraints (see Table 2). Table 2 summarises constraints reported by politicians with various strategies selection. Nominal. Politicians who cannot perceive any external affordances and only perceived internal affordance of information storing chose to have minimal presence on social media due to its impracticability for eParticipation where social media exposure is still low in the electoral base. Nominal strategy explains the minimalist approach of politicians' involvement in using social media for a two-way communication with the constituents. Politicians in this category may still perceive internal social media affordance as a tool for communication between fellow politicians. From politicians' point of view, they do not perceive the existence of external affordances of social media due to many contextual issues such as ICT infrastructure and cultural characteristic [5]. First constraint is poor Internet connection in the electoral base. This is particularly evident in the case of politicians who came from a fairly secluded electoral base such as North Maluku and Gunungkidul. In line with that, having observed two of national and district politician's web blogs, web traffic from North Maluku and from Gunungkidul is relatively low. Politicians use web blogs only to document activities for personal use such as making a personal report or preparation of plenary meeting regardless the unintended effects of doing such thing, which may induce the emergence of external affordance. In addition, politicians use WhatsApp for storing information such as political views and summary of pre-meeting discussion for effectiveness reason. Second, limited capabilities such as lack of time and lack of technical skills also hinder politicians from perceiving external affordances [9]. Third, low ICT literacy in the electoral base also gives rise to the non-use of social media for eParticipation. Evidently, all politicians at district parliament perceived that ICT literacy in the constituents is still an issue and therefore they chose conventional ways such as direct public hearing and direct meeting for opinion gathering.
Instrumental. The term describes politicians who use social media for eParticipation purpose with citizens, however due to some circumstances the politicians hire an administrator for their social media. Politicians in this stream are incapable perceiving internal affordances due to indirect relationship with social media; instead they perceive only external affordances of value sharing, opinion gathering, and promoting participation indirectly via the administrator. Lack of time is one of the reason politicians chose instrumental strategy. At first politicians handled social media accounts directly, however once they become busy and overwhelmed with the work, social media activities disrupted their workflows and therefore they decided to hire administrators. In addition, communication skill may also hinder politicians to handle social media directly, especially micro-blogging site like Twitter, which has character limitation for each posting. According to a national parliamentarian, "He (the administrator) knows my mind-set, my perspective on politics, democracy, and economics, so we communicate then he will post according to my direction, the content is my responsibility but the language is all his (responsibility)". One key factor in instrumental strategy is that the content posted on social media is the extension of the attitude of the politicians. Despite all of the reasons above, politicians still want to keep strong presence on social media as there is a demand from the constituents, therefore hiring administrator is the solution.
Manipulative. Politicians have to be cautious of their social media post since open platforms could leave politicians vulnerable to criticism from within the ranks of their constituents [10]. According to our study, the main objective of social media strategist is to manage social media content include first, to make positive image of the politicians. Second, to make politicians look capable in their own field or according to what kind of self-presentation the politicians wanted to be. Lastly, strategist usually makes provocative issues merely to boost politicians rating on social media. Not uncommon strategists were demanded to issue unusual, unique, and sometimes misleading maneuver that the public would never thought of [9]. This is done to approach segmented target on social media for instance, by paying Internet buzzer, key opinion leader and celebrities to endorse politicians with fabricated comment made by the strategist. Hence, we label this maneuver as manipulative strategy. Politicians following this strategy usually have every moment scripted and always pay attention and know when to speak up on social media. Politician perceived the affordance of personal branding highly in this strategy. Arguably because social media affords politicians to tailor the content of social media and enable them to craft and re-crafting content before it goes public [7]. Besides personal branding, affordance of value shar-ing, opinion gathering, political networking, mobilizing supporters, and promoting participation are also perceived by politicians in this group indirectly.
Genuine. We label the strategy of managing social media directly to interact with fellow politicians and constituents without using intermediary (e.q social media strategist, administrator) as genuine. The data reveal that politicians who personally managed social media account are likely to have more political connection with fellow politicians since associations are most often conceptualized by actor-initiated action [7]. Politicians following this stream perceived most of internal affordances such as idea storming, information sharing, peer entertaining and sending event invitation. For instance, politicians within the same party or commission created a WhatsApp group for information sharing and idea storming to keep the connection close. Supporting that, using social media for peer-entertaining by sharing funny content and jokes through WhatsApp group or Twitter may produce a more fluid relationship between politicians and between politicians and citizens [9].To perceive internal affordances as mentioned requires direct hands on from politicians and it would not be accomplished if intermediaries were hired for this purpose. We found that politicians at the provincial and district level are either following the nominal strategy or genuine. There is no urgency to employ intermediary since social media penetration is low to none at the provincial and district level.

Discussion
Practically, from the eParticipation perspective, the chosen strategies will have impact on the level of citizen engagement. This at least can be approached from three levels of relationship quality between politicians and citizens: information, consultation, and active participation [16]. In the context of this study, information relationship is oneway where politicians produces and delivers information to citizens. Next, consultation happens through a two-way communication in which citizens are invited to give feedback to politicians, and active participation may manifest where citizens are actively engaged in defining the policy-making process although the final decision rests on the politicians. Politicians who chose nominal strategy cannot optimally develop relationship with citizens through social media. Those in this group are neither incapable nor reluctant in using social media but rather because there is no demand from the constituents. In this case, politicians may or may not perceive internal affordances of information storing, idea storming, information sharing, peer-entertaining, and sending event invitation but when the citizens cannot afford all social media affordances then the consultation and active participation aspects would not be accomplished. The information aspect of eParticipation could still be delivered as a consequence of unintended effect of perceiving internal affordance. For instance, when politicians use social media to store information in their blogs, it creates affordance effect of value sharing to the public regardless the intention of social media use in the first place, whether to store information for personal use or to archive activities. Social media affordances at the district and provincial level is mainly perceived as internal affordance where af-fordance emerged from practice and perceived only between politicians, not to the external actors or constituents [9]. In the nominal category, citizen could not perceive social media affordances for two reasons; first, owing to the characteristic of citizens, local communities at the provincial level are still conventional even though Internet exposure is widespread. Following the customary, provincial level constituents chose to have a direct public hearing over communication via social media. Second, politicians at the district level do not perceive functional affordances due to low ICT literacy and low Internet exposure among citizens. Based on the data from the local government 2 , Internet users are minority; with only 10% of the total population have access to the Internet. These perceived constraints serve as the reference for politicians to choose the appropriate strategy in certain context [19]. To develop a better relationship with citizens, politicians use other channels, such as face-to-face physical meetings.
Both politicians and citizens' side perceived internal and external affordances when both are capable affording all social media functional affordances. This is evident in genuine strategy for instance, all identified internal and external affordances in our research include idea storming, information sharing, information storing, peerentertaining, invitation of events, value sharing, opinion gathering, political networking, personal branding, maintaining political existence, promoting participation and mobilizing supporters are related to association. Identified affordances contribute to the connection establishment between individuals and between individuals and content. The possibility to create social tie is high when politicians afford political networking via social media by connecting directly with fellow politicians and citizens as well. Other form of association is of individual with social media content that they created or recognized. For instance, information and value sharing on social media are affording relationship between the contributors of the post (e.g. politicians, citizens, social media strategist) and the audiences (social tie) plus between contributors and content. The use of hashtag and mention feature to show content reuse of the original contributors also encourage association of individuals and a piece of information as well as association between individuals [7]. Association afforded by social media is leading to participation due to the ability to forge new relationship of people whom they knew little. This stream is able to deliver the three aspects of eParticipation, information, active participation, and consultation.
Example from a social media strategist informed that a location-based social media developed exclusively to certain party may strengthen association between politicians within the same party since usually one party has widespread members across country. The location-based social media may help politicians and member of party to recognize colleague in the same area within the same party. The study also found that association in social media increased social connection for eParticipation actors by simplify easy interaction among actors. Association afforded by politicians at the national and provincial level is both social tie between politicians, politicians and citizens, and politicians and social media content. However, at the district level of parliament, politicians only afford association of individuals to a piece of information 10 not the social relationship. This is due to the nature of social media communication at the district level is one way, from politicians to citizens with minimum to none feedback.
Politicians in instrumental and manipulative strategies can't fully perceive all internal affordances due to the indirect engagement with social media. Instead they afford only external affordances indebted to a high demand from citizens. The characteristics of constituents in both strategies are ready in terms of ICT literacy, education, and Internet exposure. The demand is explained by the active participation of citizens to do a two-way consultation with politicians in many issues notably after the politicians shared information via social media. Politicians' perception of the editability of social media is exceptionally high particularly in the manipulative strategy. Editability gives opportunities for politicians to hire social media strategist to "make over" their online appearance [10]. Editability of social media allows users to strategically manipulate the ways that personal information is shared with others [7]. The interview unfold that a unified online presence of political campaign is important thus editability feature on social media is necessary. By doing this, campaign delivered the same messages the supporters wanted in the format they wanted it. The challenge towards content uniformity, however, is apparent. From the interviews, two political strategists are cautious that allowing politicians to interact directly with the constituents via social media may result in embarrassing missteps that can cost them the campaign [10].
The difference between instrumental and manipulative from the standpoint of politicians is assertive, content in instrumental is the extension of genuine idea and opinion of the politicians whilst in the manipulative usually is pre-designed and crafted by social media strategist. On top of that, as explained in instrumental strategy, lack of capability (time and skills) of the politicians is the main driver to hire administrator whereas politicians following manipulative stream sometimes are capable enough to use social media but want to stitch up the presence excellently according to their goals. Through citizens' point of view, there is no obvious difference however some active youth questioned the reliability of online content by politicians, some are even aware that social media content may be manipulated [9]. In both strategies, the deliverability of information, active participation, and consultation aspects of eParticipation is highly visible since both constituents and politicians are active on social media.

Conclusions
Building on our previous work, in this article we have identified social media strategies in eParticipation exemplified by several eParticipation actors include politicians, citizens, and social media strategists. By doing so, we have contributed to the area of eParticipation by unveiling strategies of social media use among politicians. Identified strategies are nominal, instrumental, manipulative, and genuine. As strategy may manifest in various forms (plan, ploy, pattern, position, or perspective [24]), it is then not easy to conclude whether the perceived internal and external affordances will lead to a certain selection of strategy or another way around. Several constraints are also identified during the data analysis and are influencing the choice of strategies. Identified constraints include poor Internet connection, limited capabilities, characteristic of citizens, low ICT literacy, and unreliable social media content. Each of the strategies is affected by different motivations and constraints of social media use.
Our study offered two main contributions. Firstly, practical contribution identifies strategies of social media use that may advise politicians in formulating eParticipation include nominal, instrumental, manipulative, and genuine. Secondly, conceptualization of internal and external affordances from the perspective of politicians and citizens projected by eParticipation concept of information, consultation, and active participation allows us to understand how social media affordances encourage politicians' choice of strategies to promote citizens participation. By so doing, we have contributed, albeit minor, to the theory of social media affordances by demonstrating that politicians who perceived affordances-for-practise are also motivated by their perception of functional affordances offered by social media. Hence, not only could social media functional affordances bring changes to the way that many processes are carried out in organizational context [7], but also it can bring changes to many processes carried out by individual.
We have argued that the presence of social media technology may help politicians in delivering information, encouraging active participation as well as providing a twoway consultation between politicians and constituents depending on the choice of strategies. First, nominal strategy explains the minimalist presence of politicians on social media due to several constraints such as low ICT literacy and poor Internet connection. Second, instrumental strategy describes politicians who hire social media administrator due to their lack of capabilities in using social media. Third, manipulative explains the strategy to employ strategist to make up social media appearance in accordance with certain goals by creating provocative political maneuver. Lastly, genuine strategy describes the way in which social media are managed and used by the politicians for eParticipation without any intermediaries. Moreover, herein, citizens played an important role in this scene; citizens' perception of social media affordance may encourage or discourage politicians in using social media.
We provide several future avenues, first, since our study does not discuss constraints in any depth, but only acknowledging constraints as factors that may prevent actors from perceiving affordances, further investigation may discuss how constraints can be overcome in order to better utilize social media in eParticipation. Second, our study has not examined further on the possibility of choosing more than one strategy for eParticipation by the politicians. Future study may investigate the choice of strategy and examine how and when certain strategy is better than others in a particular time such as before or after election.