Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Political parties can potentially benefit from Social Media such as Facebook, Twitter, YouTube, Google+, and LinkedIn to improve interactions between their members. For example, multiple studies have indicated that the Social Media strategies of Howard Dean, Barack Obama, and Ségolène Royal contributed to members becoming more engaged (Christakis and Fowler 2009; Citron 2010; Greengard 2009; Lilleker et al. 2010; Montero 2009; Talbot 2008; Ren and Meister 2010; Zhang et al. 2010). In the case of Obama, the members with higher engagement donated more to the party and also were more willing to take an active part in the campaign. In the case of Royal, party membership increased from 120,000 to 200,000 members, 90 % of whom had not previously been members of a political party (Montero 2009).

The Arab revolutions of 2011 are other examples of the impact of Social Media. During the “Arab Spring”, voices of normally ignored people could reach and influence people all over the world (Howard and Hussain 2011). These examples show that Social Media can affect party politics and democracy more generally. However, we know little of precisely how—and to what extent—Social Media participation affects politics. Why do certain politicians benefit from Social Media, while others do not? Are social network sites purely reflecting preexisting offline social networks? Yet, these questions remain unanswered.

Society changes with the expansion of science and technology (Latour 2005). Social Media, as products of new technology, have a high impact on society. The Internet has become increasingly social. As of January 2012, Facebook has 800 million registered users, and according to market researcher ComScore (Ray 2010), people are spending even more time on the Facebook than on the Google. The increased use of the mobile Internet by users of smartphones and tablet computers contributed significantly to the adoption and use of Social Media. In the United States, Social Media reach nearly 80 % of active Internet users and currently represents the majority of Americans’ time online (Nielsen 2011). In Western Europe, these numbers are even higher. Increasingly, people are connected to each other, without regard to time or place.

As people and politicians increasingly adopt Social Media, measuring the effects of Social Media Participation on party community participation has become more important. However, to our knowledge, effective evaluation methods remain lacking. A systematic literature review that we recently conducted revealed that there is a lack of measurement instruments and most existing instruments in the e-participation field are not capable to evaluate the effects of Social Media. (Effing et al. 2011). Our survey further revealed that the available instruments primarily focus on pre-Social Media Internet tools such as forums, chat, and online surveys (Phang and Kankanhalli 2008; Roeder et al. 2005; Stern et al. 2009). Only a few frameworks are capable of evaluating Social Media participation, such as the e-participation ladder of Macintosh (Grönlund 2009). However, these frameworks are too high level in perspective and not ready to evaluate Social Media participation directly from the available empirical data.

In this chapter, a measurement model is proposed that will be able to measure the community effects of Social Media. Improvements in measurement can guide researchers and politicians about which Social Media to use and which strategies are the most effective.

Therefore, the main question addressed in this chapter is: What determines Social Media Participation and how can the effects on political parties be best measured?

Let us, first define the main elements of this question. Political party communities are relational communities for a professional cause and are not necessarily territorially bounded (McMillan and Chavis 1986). The members of political parties are engaged in their communities because of shared beliefs, goals, or interests.

Grönlund (2009) defines participation as “the specific activity of doing things together”. Xie and Jaeger (2008) define political participation as “behaviors aimed at shaping governmental policy, either by influencing the selection of government personnel or by affecting their choices”. Participation is doing things together for a shared belief that government policy should change in the parties’ direction. Participation is one of the key elements of Social Media. Kaplan and Haenlein (2010, p. 61) define Social Media as: “a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content.” This definition makes clear that Social Media as a term is not a completely new generation of Internet tools. Social Media rely heavily on the concept of Web 2.0. “Web 2.0 is a term that was first used in 2004 to describe a new way in which software developers and end-users started to utilize the World Wide Web; that is, as a platform whereby content and applications are no longer created and published by individuals, but instead are continuously modified by all users in a participatory and collaborative fashion” (Kaplan and Haenlein 2010, p. 60). People collaborate in communities and the Internet evolved into a place where many people have collaborative tools at their fingertips.

To measure the effects of Social Media Participation on Political Communities, we propose a conceptual measurement model based on two concepts: Social Media Participation and Community Participation. To measure Social Media Participation, we developed a standardized instrument termed the Social Media Indicator to assess the use of Social Media by politicians; a set of questions to assess the degree of Social Media participation based on e-participation theory by Macintosh and Smith (2002). Additionally, we integrate established media-choice-theory from Short et al. (1976); Rice (1993), and Te’eni (2001) to include the aspects of the choice and appropriateness of Social Media. For measuring Community Participation, we deploy the following two constructs: Community Engagement (CE) and Sense of Community (SOC) (McMillan and Chavis 1986).

The proposed conceptual measurement model is shown in Fig. 13.1.

Fig. 13.1
figure 1

Social media participation model

Full size image

Using this model, we aim to discover if, how and to what extent certain Social Media strategies affect the participation of communities’ members. This relationship is visualized by the arrow between the concepts of Social Media Participation and Community Participation. This model is a major simplification of the empirical situation. But, as made clear by Blumberg et al. (2011, p. 155), causal studies “cannot observe and measure all the processes that may account for the A-B relationship”. While more complex models could be designed, this first version focuses on key factors to measure effects.

We hypothesize that a causal relationship exists between the two concepts. Our grounds to assume this are three-fold:

  1. 1.

    The number of relationships between people tends to increase when people use social network sites, because these sites reveal relationships by making them visible (Boyd and Ellison 2008). Consequently, people connect more easily with each other. For example, a study of Tomai et al. (2010) showed that members of a virtual community of a school had significantly higher levels of bridging social capital.

  2. 2.

    Online behavior changes offline behavior. For instance, the use of Social Media can reduce transaction costs of communication (Ren and Meister 2010). Consequently, this can lead to other choices in communication channels for certain tasks. While online communication can replace certain forms of offline communication, studies show that online relationships do not replace offline relationships, but augment them (Vergeer and Pelzer 2009).

  3. 3.

    If people are connected online, this contributes to general feelings of attachment to the community: the Sense of Community. People are confronted with their connection to the community at other places and other times, which makes them aware of their existing relationships more frequently. Consequently, this can increase their feeling of being part of the community in general (Tomai et al. 2010). However, we should be aware that an already existing Sense of Community can influence Social Media use. In that case, offline relationships are, partly, mirrored in online relationships and will stimulate the use of Social Media. Therefore, the arrow could also be drawn in both directions.

Given the reasons above, Social Media Participation is assumed to influence Community Participation overtime. By using this model in empirical studies, future outcomes will provide evidence to accept or reject this hypothesis.

The remainder of this chapter is structured as follows: In the next section, we will define Social Media Participation in more detail by operationalizing the constructs of the model. In the third section, we will do the same for Community Participation. In the fourth section, we propose directions for applying the Social Media Participation Model. In the final section, we discuss the study, draw conclusions, and make recommendations for future work.

2 Social Media Participation

In this section, we propose a method for the measurement of Social Media Participation by introducing operationalization for the constructs of Social Media Use and Social Media Choice. However, we will first explain underlying theory of Social Media Use. The concept of Social Media Use refers to the upper left construct of our model.

2.1 Social Media Use

Effective measurement instruments must be able to produce detailed data to evaluate and compare Social Media use of individual politicians. This is the key reason why we developed our own instrument.

The Social Media Indicator (SMI) evaluates the use of Social Media by politicians. The indicator combines Contribution (sending information and content) and Interaction (discussion, dialog). The indicator comprises a set of standardized questions that will deliver scores that indicate the extent to which individual politicians are using Social Media. The scores can be used to indicate adoption levels of Social Media, but can also be used to assess both levels of contribution and interaction. Therefore, in addition to total SMI scores, the instrument provides scores for Contribution and Interaction. Currently, the instrument measures the following Social Media tools: Weblog (Blog), Facebook, YouTube, Twitter, and LinkedIn. Overtime, the instrument will be extended with other Social Media tools as well.

The standardized questions of the Social Media Indicator are presented in Table 13.1.

Table 13.1 Social media indicator
Full size table

After answering the SMI questions, it is possible to calculate and compare personal SMI scores for each member of a political party. Every act of communication represents one point, because we decided that every person-to-person interaction counts the same, regardless of the medium. A time interval must be defined before collecting the data. Overtime, measurements must be repeated to see how participation develops. For the first measurement, it can be useful to calculate the score on the basis of the entire history of Social Media use by the politicians.

We claim that the SMI that we have devised makes the participation levels of politicians in Social Media both visible and comparable. Three reasons underpin both reliability and validity of the SMI: (1) an empirical example; (2) solid underlying theory; and (3) public accessibility of data. We will describe each of these next.

First, there is an empirical example available where the SMI has been applied. This is the case of the elections in the Netherlands. The instrument was tested at the national level on all Dutch political parties and all the candidates for the Second Chamber election and it was able to collect data that was used for calculation of statistically significant correlations (Effing et al. 2011).

In the example above, the relation between Social Media and voting outcome were measured by the SMI. Therefore, politician’s personal SMI scores were compared with personal votes. Scatterplot diagrams and the calculation of Spearman’s rank correlations revealed the following outcomes:

Within 9 parties, out of 16, a positive significant correlation was found between SMI and votes as illustrated in Table 13.2.

Table 13.2 Correlations between SMI and voting outcome within the Netherlands
Full size table

The differences in correlations could be a result of differences in target audience, content strategy, and other factors, but these factors are not yet thoroughly explored.

Although the empirical results show that Social Media Use in 9 out of 16 cases has a positive relationship with voting outcome, we could not completely explain what determined the relationship (Effing et al. 2011). This emphasized the need for further research. Also, for six parties, this relationship was not significant. To understand what determines effectiveness in Social Media use, additional interviews with Dutch political parties revealed that the presence of underlying strategies partly determine the variations in the effectiveness of Social Media. Target group differences could also partly explain variations in significance of correlations.

Secondly, the SMI is grounded in established theory. The instrument is based on the frequently cited participation ladder of Macintosh (Grönlund 2009; Medaglia 2007; Sommer and Cullen 2009). Macintosh created a three-step participation ladder, which is useful for describing the participation levels of the Social Media phenomenon at a high level. Other e-participation ladders from the literature might also be useful, but we found Macintosh’s model to be most suitable for Social Media. The first step in the ladder is e-Enabling. In this step, party members provide access and information to citizens. The second step is e-Engaging. During this stage, party members give opportunities to citizens to interact with them and start a dialogue. Citizens are frequently consulted on certain projects, decisions, or activities, for instance through forums and polls. The third step is e-Empowering. This step is about members working together with citizens, empowering the citizens with responsibilities, tasks, and options to collaborate with the party’s community. Previous efforts at trying to empower citizens often failed (Phang and Kankanhalli 2008; Roeder et al. 2005; Stern et al. 2009). This was due to immature technology and low user adoption rates. As Social Media mature, the challenge remains to discover how Social Media can accomplish e-Empowering. However, e-Empowering is not directly recognizable from the SMI data without additional inquiry. For this reason, we argue that Social Media choice aspects should also be part of the analysis of Social Media participation.

Thirdly, most of the data necessary to calculate the SMI scores are available from open databases. Although certain statistics are not accessible due to privacy settings, the majority of personal data from Social Network Sites are publicly accessible (Boyd and Hargittai 2010). In cases, where authorization is required to access required data, collaboration with parties can be the answer. However, since most of political communication is public debate, this is not a key problem.

In our projects, we listed the top five Social Media, which were representative of the vast majority of all Social Media traffic, based on numbers of advertisement-reach from market researchers. In the Netherlands, for instance, Hyves is one of the largest Social Network Sites, therefore, it should be part of the Social Media Indicator to obtain valid results (Comscore 2011).

Because the SMI score is only an indicator, it is unnecessary and impractical to include all available Social Media tools. Social Media tools with low adoption rates are not included because of their low reach. However, there could be specific reasons to include Social Media that are less common. For instance, political party communities could use the internal Social Medium called Yammer. In that case, this medium may be included. Indicators should be investigated carefully, before being included into the SMI. For instance, the view count of Hyves in the Netherlands is not a valid indicator of participation because artificial users such as search engine spiders heavily skew the results. In such a case, the total apparent score is biased.

Our experience with the SMI has demonstrated that the Social Media Indicator is an effective method for measuring and comparing the degree of use and participation but does not fully explain the differences in effectiveness.

2.2 Social Media Choice

“Nothing impacts the success of a Social Media effort more than the choice of its purpose.” (Bradley and McDonald 2011) Social Media tools are not effective for everything. For some politicians, participating in Social Media seems to be a goal in itself. Not all communication by Social Media is appropriate for all communication strategies. According to Te’eni (2001, p. 1), “current technology can affect not only how we communicate but also what we communicate.” Therefore, we have to take the choice and appropriateness of Social Media into account when determining variations in impact on the dependent concept of Community Participation. This part is illustrated in the left bottom construct of our model (Fig. 13.1). According to Rice (1993, p. 453), appropriateness is “a good match between the characteristics of a medium and one’s communication activities”.

From our open interviews with Dutch political parties in 2010, we learned that one of the explanatory factors for possible influences were the underlying communication strategies (Effing et al. 2011).

Furthermore, in the cases of Obama, Dean, and Royal, all of them thought thoroughly about both choice of and strategy for using various Social Media for different purposes and target groups (Christakis and Fowler 2009; Citron 2010; Greengard 2009; Lilleker et al. 2010; Montero 2009; Talbot 2008; Ren and Meister 2010; Zhang et al. 2010). The combination of our recent studies and the experiences of Obama, Dean, and Royal make a strong case to include Social Media choice aspects in our measurement model.

An extensive body of literature is available from the communication and Information Systems fields, which focuses on the choice, capacities, strategies, and appropriateness of media. After a literature review, based on the relevance and frequency of citations, we selected the theories of Social Presence (Short et al. 1976), Media Appropriateness (Daft and Lengel 1986; Rice 1993), and the Theory of Cognitive and Affective Organizational Communication (Te’eni 2001) to strengthen our model. Although all of those theories have certain shortcomings, they provide a helpful theoretical background for understanding differences in the appropriateness of Social Media. The theories of Social Presence and Media Appropriateness should not be applied too strictly, because users cope effectively with the limitations of digital communication and they invent ways to transmit social cues through these media (Morris and Ogan 1996).

In the selected theories, various communication strategies are presented.

Short et al. (1976), present the following communication activities.

  • Exchanging information.

  • Problem solving and making decisions.

  • Exchanging opinions.

  • Generating ideas.

  • Persuasion.

  • Getting the other on one’s side of an argument.

  • Resolving disagreements or conflicts.

  • Maintaining friendly relations/staying in touch.

  • Bargaining.

  • Getting to know someone.

In addition, Te’eni (2001) presents the following communication strategies:

  • Contextualization (how and why and meta-data): “provision of explicit context in the message” to increase comprehension.

  • Affectivity: inclusion of affective components in the message that describe emotions and moods.

  • Control [by testing/planning]: redundancy and repeated communication. Timely feedback is essential for effective control.

  • Perspective taking: actively considering the receiver’s point of view, inquiring of them about their affairs and attitudes and supporting them, sharing common beliefs and talking in a personal style.

  • Attention focusing: manipulating the receiver’s processing of the message by emphasizing (switching style, highlighting, shouting, pervasive techniques).

Media differ in terms of efficiency and capability of reaching the desired outcome of communication strategies. According to Short et al. (1976), media vary in Social Presence, which is “the degree to which a medium is perceived as conveying the presence of the communicating participants”. Social Media differ in terms of capacity to transfer Social Presence.

Apart from Social Presence, media also differ in terms of interaction level. In regards to different forms of media, differences exist in their capacity to handle immediate feedback from the communicating participants and differences in the way social cues can be part of the communication. According to Daft and Lengel (1986, p. 560), “media vary in capacity to process rich information”. Information richness is defined as: “the ability of information to change understanding within a time interval” (Daft and Lengel 1986, p. 560). In that sense, the amount of time required for a medium to provide understanding is also considered an important element when considering the richness of a medium.

Rice (1993) was one of the first to apply the theories of Social Presence and Media Richness to new media. Based on the empirical evaluation of Rice (1993), people perceive the following hierarchy in appropriateness of media for communication activities of Short, Williams, and Christie:

  1. 1.

    Face-to-face (most appropriate for 8 out of 10 activities).

  2. 2.

    Phone (most appropriate for time-sensitive information).

  3. 3.

    Meeting (scheduling/organizing, temporal, and physical obstacles.

  4. 4.

    Desktop video.

  5. 5.

    Vmail.

  6. 6.

    Text.

  7. 7.

    E-Mail/(New media) (appropriate for exchanging information and time-sensitive information, asking questions, staying in touch).

Types of Social Media, as specific forms of new media, have different levels of appropriateness for different communication strategies (Kaplan and Haenlein 2010).

To develop a classification scheme for our measurements, we now present our Social Media Appropriateness Matrix. On the left vertical axis, we present the degree to which Social Media have the ability to facilitate direct feedback and interactivity. Social Media differ in the potential time it takes to receive immediate feedback. On the right horizontal axis, we present the extent to which Social Media can be perceived as personal. This axis relates to Social Presence. In Fig. 13.2, we present our Social Media Appropriateness Matrix, which compares different types of Social Media from the perspective of appropriateness for communication. The labels within the matrix are partly based on the definitions given by Kaplan and Haenlein (2010). They do not reflect actual perceived levels from politicians and have not yet been empirically tested.

Fig. 13.2
figure 2

Social media appropriateness matrix

Full size image

Social Presence mediated by Social Media is lower than, for instance, face-to-face conversations. However, as Fig. 13.2 shows, videoconferencing with a high-speed connection via Skype can be perceived as a form of higher Social Presence. This is because social cues are made visible through webcam video and voice transmission. Twitter, being mainly a text message micro-blogging system, is assumed to have a lower Social Presence than, for instance, a personal YouTube web video, but has the ability for immediate feedback by followers. Therefore, we classify micro-blogging as an example of higher interactivity. Nevertheless, this feedback level is lower than in face-to-face or telephone communication. Social Network Sites, such as Facebook, offer opportunities to generate a certain level of Social Presence. Social Presence is increased due to the creation of personal profiles with interests, maintaining networks of relationships, and sharing personal content, such as pictures.

This Social Media Appropriateness matrix, in combination with communication strategies, provides a classification scheme to evaluate Social Media strategies in structured, semi-open interviews. Repeating those interviews overtime will provide results about changing strategies overtime.

At this point, we have discussed all relevant constructs of the left part of the SMPM (Fig. 13.1): Social Media Participation. Now, we will elaborate on the concept of Community Participation.

3 Community Participation

We divide the concept Community Participation into two constructs. The first construct is Community Engagement. The second construct is Sense of Community. We explain each of these next.

3.1 Community Engagement

To evaluate the level of general offline and online community engagement, we collect data about individual members’ activity within the community. Community Engagement represents the upper right construct in our model (Fig. 13.1). Community participation addresses more than the online environment alone. Data collection of one or more of the following indicators at the overarching level of community is required:

  • Time spent.

  • Presence at meetings.

  • Money donated.

  • Number of Legislature Bills, requests, or ideas contributed.

  • Other activity indicators.

Asking survey questions to retrieve these data is possible, but could deliver biased results. The reason is that asking respondents directly will influence them, because they then become aware of their (lack of) engagement. Unobtrusive methods are preferable to obtain similar data, such as document or database analysis.

Next to the data mentioned above, basic social networking analysis can provide data about how community members are interconnected. For example, the community may be formed around one powerful leader, while in other cases, the community power is distributed among many politicians. Political friends tend to influence each other. Christakis and Fowler (2009) showed that being connected to each other in a social network influences political party campaigns, voting and cosponsorship within politics. Making basic network diagrams of a party network can help to understand how communication, power, and influence within a party are distributed. In addition, subgroups, powerful leaders, or disengaged members tend to become visible. The investigation of those elements and networks can be repeated overtime to see how the community engagement develops or collapses.

Community Participation also relies on softer factors such as Sense of Community, which will be discussed below.

3.2 Sense of Community

Measurement of participation within political communities involves more than measuring Community Engagement. The Sense of Community largely covers the psychological aspect of Community Participation. This construct is visualized in the bottom right part of our model (Fig. 13.1).

Chavis especially contributed to the scholarly literature in regard to Sense of Community (McMillan and Chavis 1986; Chavis and Pretty 1999; Chavis et al. 2008). The concept of Sense of Community has been used in numerous social studies (Chavis et al. 2008). Sense of Community (SOC): “is a feeling that members have of belonging, a feeling that members matter to one another and to the group and a shared faith that members’ needs will be met through their commitment to be together.” (McMillan and Chavis 1986, p. 9)

It consists of four elements (McMillan and Chavis 1986, p. 9):

  1. 1.

    Membership: “the feeling of belonging or of sharing a sense of personal relatedness.”

  2. 2.

    Influence: “a sense of mattering, of making a difference to a group and of the group mattering to its members.”

  3. 3.

    Reinforcement/integration and fulfillment of needs: “the feeling that members’ needs will be met by the resources received through their membership of the group.”

  4. 4.

    Shared emotional connection: “the commitment and belief that members have shared and will share history, common places, time together and similar experiences.”

It is difficult to view the four elements in isolation because the elements influence each other.

Examples of Sense of Community studies are broadly available in the literature to explain the dynamics within various communities, such as neighborhoods, youth gangs, kibbutz, churches, workplaces, schools, universities, recreational clubs, and Internet communities. The Sense of Community theory does not limit itself to a certain type of community and is therefore useful to describe and compare various types of community. McMillan and Chavis (1986), p. 19 argue that “because of their common core, although our four elements will be of varying importance depending on the particular community and its membership. These elements, then, can provide a framework for comparing and contrasting various communities”.

The last few decades have seen frequent testing and refinements in measuring Sense of Community. “Researchers do not appear ready to settle on a definitive and consistent SOC measure” (Chavis and Pretty 1999, p. 636). One of the most refined measurement instruments is the so-called SCI-2 (Chavis et al. 2008). It consists of 24 statements that individuals can respond to on a Likert scale. The SCI-2 was used in a survey of 1,800 people and the measure’s reliability was found to be very high (coefficient alpha = 0.94). This SCI-2 instrument is effective in evaluating how strongly members feel attached to their political party’s community. More importantly, it is possible to measure how community attachment develops overtime, if measurements are repeated.

Now that the measurement model for measuring Social Media and Community Participation has been described, we will propose a method to apply the model in future research projects.

4 Application of the Social Media Participation Model

Based on the Social Media Participation Model, it is possible to design a causal study to reveal relationships between Social Media Participation and Community Participation. For example, to discover how the use of Social Media affects the community of a political party at a local municipality. In this section, we propose guidelines for using the model.

The proposed guidelines for studies applying the Social Media Participation Model are based on comparative case study research (Yin 2008), including both quantitative and qualitative data collection techniques. According to Waters et al. (2009), “longitudinal studies could offer insights into how organizations change their social networking strategies overtime, and case studies should be conducted to help offer insights for other organizations based on efforts that have both succeeded and failed”.

Table 13.3 summarizes guidelines for applying the model and underlying measurement constructs, which can be part of longitudinal case studies.

Table 13.3 Suggestions to apply the social media participation model
Full size table

Constructs can be related to each other. SMI scores can be compared with both levels of engagement as the number of Sense of Community. This is particularly important when political parties increase their Social Media activities. When measurements are repeated overtime, they will provide insights into how communities—such as political parties—change by Social Media.

Because multiple influence factors are involved in complex community settings, we suggest a combination of quantitative and qualitative research methods. However, the model can be used for a variety of research designs.

5 Conclusion and Discussion

In this chapter, we have aimed to create an understanding of how Social Media Participation affects political party communities, and proposed the Social Media Participation Model (SMPM) as illustrated in Fig. 13.1. In the final section, we will conclude the chapter with a short summary, describe the limitations of the proposed model, and give pointers for future research.

5.1 Conclusion

Which forms of Social Media participation influence aspects of the Community Participation the most? By getting results from the SMPM, it will be clear which choice and use of Social Media positively influence member participation within political communities. Our hypothesis is that Social Media participation, using certain strategies and with appropriate media selection, can increase Community Participation in political party communities.

We proposed the SMPM to help understand how Social Media Participation affects community participation. We combined four measurement constructs from theory and practice into one integrated model.

This model is a first step in developing a standardized instrument to compare Social Media Participation with Community Participation. Although the model mainly consists of constructs used in established theories, the model still needs further improvement and empirical testing.

5.2 Discussion

The advantages of measuring with the SMPM are that data analysis can be carried out to compare left and right concepts from the model. For instance, correlations can be calculated between SMI scores and SCI scores. Also, we can make different comparisons for different communication strategies (Social Media Appropriateness) and analyze which strategies affect Community Participation the most.

Nevertheless, several limitations exist for our measurement model and its constructs.

First, obtaining results from the Social Media Indicator by observation is still time consuming since appropriate integrated social listening tools are still lacking or are being developed. Currently, social-listening tools, such as Radian6 and Teezir, cover only parts of the necessary data. A second limitation is that due to privacy control, not all Social Media participation data are publicly accessible.

A second limitation is that Sense of Community can consist of various echelons (such as local versus national communities) and therefore can be complex to measure if boundaries between communities are not clear.

The SMPM is based on a linear causal view, while in reality the constructs also influence each other in cycles. This is the third limitation.

A final limitation, presented here, is that not all politicians are willing to use Social Media for various reasons. As addressed earlier in this chapter, Social Media does not replace other channels of interaction but augment them. The Social Media Participation model does focus entirely on the effects of Social Media and does not cover other instruments for interacting with and between politicians that could cause changes in Community Participation as well.

5.3 Future research

Future results obtained by using this model should bring us further knowledge about which Social Media strategies are most effective for political party communities, and prove the reliability and validity of the model. By applying the proposed model and methods, we have designed a longitudinal study at the council of the municipality of Enschede, which has more than 160,000 citizens and is located in the eastern part of the Netherlands. We plan to conduct a series of tests based on this measurement model. They will show us how and to what extent Social Media changes communities operating within Enschede’s city council. It is hoped that design principles for effective Social Media implementations, can be derived from the empirical results.

This measurement model is part of a larger research project. In this project, we also investigate other types of communities, such as churches. With a broad selection of cases of the use of social media in not-for-profit communities, interesting comparisons can be made. With future outcomes, relevant advice can be given to political parties about how to improve their interaction with their communities by using Social Media.

Next to our own future studies, we also encourage other scholars to test and refine this research model. A more refined cause-and-effect model could help to increase understanding of the effects of Social Media Participation on Community Participation. At this point, questions remain about how Social Media affects politics, all over the world.