Social media and politics: Can communication
professionals replace politicians?
A study on the effect of deployment of communication professionals on
political social media on the evaluation of messages and political actors.
Master Thesis
Kirsten van Ewijk Student number: 10002324
Adviser: Damian Trilling Date: 27/06/2014
Social media are integrated in political campaigns more and more. When politicians use these media properly, this helps to maintain a positive image, which in turn could have a significant effect on election outcomes. However, politicians do not always have the time or expertise to administer their social media account as they should to achieve this goal. Letting de communication department manage the account would be the perfect solution, was it not that the deployment of these professionals in politics has a negative connotation among the public. By conducting an Internet experiment this research examines if the deployment of communication professionals on social media pages of political parties and politicians has a negative effect on the evaluation of these parties or politicians and the messages. Therewith it is examined if this effect is different for pages of political parties and politicians. However, there were no significant results found and there are no conclusions that can be drawn.
Key words: Communication professionals, interactivity, professionalized political campaigns, political communication, social media, social presence, spin doctor.
If you compare a recent political campaign to a campaign that was held hundred years ago you will find many differences. Various societal and media changes that occurred during the twentieth century have forced politicians to campaign in a different way (Blumler & Kavanagh, 2010). New forms of media such as television were increasing and the publics’ main reason for the use of media was to be entertained. Politics was not guaranteed of a place in news media anymore and other ways had to be found to stay in the eye of the citizens. Because the public was searching for entertainment, the focus of political reporting shifted towards other aspects of the campaign. There was more emphasis on a politician’s appearance, private life and personal aspects while the political statements were far less attention given (Strömbäck, 2007). Politicians had to change their campaigns to prevent that these factors would influence the election results in a negative way.
“Professionalized political campaigns” is a term that has been used increasingly over the last century (Strömbäck, 2007). Now that other aspects became more important a politician did not mainly had to focus on his viewpoints, but it was also essential that messages were formulated properly that the right image was built up. Therefore campaign teams were extended and extern professionals were hired. Communication professionals, campaign managers and stylist started to support politicians in order to achieve the best results in election time: Get the politician to win the election. They were the ones who told the politicians where to speak, how to speak, how to dress and how to maintain the right image (Blumler & Kavanagh, 2010).
The evolvement of political campaigns did not just stop there. The media landscape changes continuously and politics has to adjust to these changes adequately. Nowadays new media entered the media landscape and have increased in popularity tremendously. In order to reach a large audience it is essential for a politician to be where the people are, and currently that is the World Wide Web. Social media are having a growing role in today’s society (Shao, 2009; Gil de Zuniga, Jung & Valenzuela, 2012). While these media started out as popular media for the younger generations, nowadays these media have gained popularity among the older generations too. In order to get its message across, practically every political party or politicians has added social media to its campaign (Crawford, 2009; Karlsen, 2010). By connecting with the audience via social media political parties and politicians have the ability to promote themselves, mobilize voters and to have interactive
communications with their audiences (Kruikemeier, 2014), something that was not as easy before the period of social media.
Now that social media have found their place in political campaigns it is essential that the content on these pages is checked and updated frequently (Kruikemeier, 2014). Especially for the personal pages of politicians it would be expected that the politicians themselves create the content for these pages. However, there are two difficulties in this aspect. Firstly, a politician’s job is not to be on social media the whole day. Secondly, there is a risk in letting politicians post whatever they want. A politician is not specialized in this form of communication and wrongly formulated messages could not only harm the politician’s image but could affect the whole party (Vergeer, Hermans & Sams, 2013). While communication professionals are already a part of campaign teams, these problems could by solved by letting them administer these social media accounts on behalf of the politicians. However, this could raise different reactions by the public, since they have the idea they are interacting with a party or politician itself.
The main purpose of this research is to investigate if the deployment of communication professionals on social media pages of political parties and politicians has an effect on the image of a party or politician and the effectiveness of the messages. An Internet experiment is held where the communicator of the social media messages was manipulated; For one group the party or politician administered the social media page, for the other this was the communication department. The outcome of this research is beneficial for society, as citizens are the audience of the politicians, and the politician themselves, since they are the ones who need to send the most effective messages across. But there is also a scientific benefit. Although there has been done a lot of research on social media and the effectiveness of messages in general, no connection has been made between the communicator of social media and its effects. This research has the potential to close this gap.
In the next section there will be explained what social media actually are and how they became popular. Then the aspects that differentiate them from traditional media will be highlighted and lastly it will be explained what social media add to political campaigns and what influence they have on the evaluation of politicians.
Theoretical framework The development of social media
At the end of the twentieth century the Internet was gaining popularity very quickly. However, Internet users were only able to read and search for specific information and there was little opportunity for them to actually add something on the World Wide Web. Over the years the Internet kept evolving and in 2004 a new milestone was reached: the term ‘Web 2.0’ was introduced (Kaplan & Haenlein, 2010). Web 2.0 is the name for “a platform whereby content and applications are no longer created and published by individuals, but are continuously modified by all the users in a participatory and collaborative fashion” (Kaplan & Haenlein, 2010, p. 61). In other words, Internet users now had the ability to create content and publish this on the Internet easily. This content is also referred to as User Generated Content, content that is generated by the users of the social media (Kaplan & Haenlein, 2010).
When researching social media it is essential to know what social media exactly are. Social media are websites and software that allow their users to connect, communicate and interact with each other. Basically there are three types of social media: Social networking sites, tools for communication, and website to share information with others (Wang, Tchernev & Solloway, 2012). While all types of social media are very popular, social networking sites are the most popular form of all. Social networking sites are “web-based services that allow individuals to (1) construct a public or semi-public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections to those made by others within the system” (Boyd & Ellison, 2008, p. 211). Based on interest and goals there is various content that could be shared. While some type of social media mainly have the tasks of sending and receiving messages to specific people, other social media allow users to share posts, pictures, or videos with all the users of the web, and via some social media it is possible to combine all of these activities (Shao, 2009).
The use and gratifications of social media
Currently there are two main trends in the use of social media (Quan-Haase & Young, 2010). Firstly, social media users do not use just one type of social medium. Mainly social media users use various kinds of social media, because every type of social medium offers different possibilities for its users. Therefore social media sites do not replace each other, but rather complement each other. Secondly, when a social medium becomes popular and increases in popularity, the usage of the former popular
social medium decreases. However, since social media complement each other the usage of this social medium does not diminish entirely (Quan-Haase & Young, 2010). This makes sense, because otherwise there would not be so many different social media sites nowadays. However, this does not mean that social media sites will never disappear, but the introduction of new social media sites does not have to mean that the former sites would not be used anymore.
The questions that remain are why are social media so popular and, even more importantly, why is it beneficial for politicians to integrate social media in their campaigns. When television became one of the most popular media a lot of research has been done on why people choose for a certain medium. According to the uses and gratifications theory the audience actively chooses a medium with a specific goal. Every medium has the ability to fill a certain need and the audience chooses a medium based on the need that they want to be fulfilled (Katz, Blumler & Gurevitch, 1973). Although traditionally a lot of research on this theory has been done for the television medium, research on the uses and gratifications theory concerning other types of media is growing.
The main purpose of social media and the main reason for the public to use these media seem to correspondent with each other. The public mainly joins social media to keep in touch with a current network of friends, family and acquaintances (Utz, 2009). This interactivity-aspect is what social media are all about. Interactivity is “the extent to which the communicator and the audience respond to, or are willing to facilitate, each other’s communication needs” (Ha & James, 1998, p. 461). Every person has different needs and therefore uses social media in its own way. In general there are three different ways to make use of social media: By consuming, participating, and producing information (Shao, 2009). Consuming information is reading and viewing the content that is published on social media without reacting or posting content itself. The only thing a user does is consuming the information published by others. When a person reacts or shares certain content or comments on posts from other users, this person is participating. The user does not publish content, but makes use of the social medium actively. Only when a person creates and publishes actual content this means that this person is producing content on social media. This type of social media use is producing information.
Based on a user’s personal needs a person uses social media certain way. However, with social media there is always a certain level of interactivity involved.
Interactivity could be divided in five dimensions, namely: (1) Playfulness, (2) choice, (3) connectedness, (4) information collection, and (5) reciprocal communication (Ha & James, 1998). The first dimension, ‘playfulness’, refers to the entertaining part of the use of social networking sites. Entertainment is considered to be the essence of interactivity. Just by clicking on certain buttons the audience could experience joy and find what he is searching for. Choice, the second dimension of interactivity, refers to all the opportunities of social media. The choice to be as engaged as you want makes consumers feel empowered. Thereby users can click on every link they want, which makes is possible to jump from one page to another. The third dimension of interactivity ‘connectedness’ focuses on this aspect. This feeling of connectedness makes users feel like they are connected to everything in the world. The fourth dimension of interactivity is ‘information collection’. Via the Internet senders of the messages have the opportunity to receive information of their audience. This makes it possible for them to adjust their messages to specific characteristics of the receivers and therewith improve the relationship. Lastly, the fifth dimension of interactivity is ‘reciprocal communication’. This is one aspect that really differentiates social media from traditional media or websites. While communication via traditional media was mainly one-way communication, now there is the opportunity for users to actually interaction with each other and have conversations.
Social media and politics
Although social media are mainly used to connect with people that are already a part of someone personal network (Utz, 2009), there are also opportunities for other aspects of society on these media. Social media are more and more adopted in marketing and political campaigns, because organizations and politicians need to appear where their audience is. By implementing social media in a campaign, the original triadic relationship between journalists, politicians and citizens could be changed (Bakker, 2013). In this original relationship politicians need journalists to get reach most of the citizens. However, they have no influence on when and how their messages are published. Journalists can choose their own frame and select specific quotes that could influence the audience. On the other side journalists need politicians to get information about important events and inside information (Van Aelst, Shehata & Van Dalen, 2010). In this original relationship the only task for citizens is to receive information and there is no room for interaction and conversation. Social
media could change this relationship in three ways (Bakker, 2013). Firstly, politicians now have the ability to publish their own messages, and therewith bypass the journalists. By doing so political messages can be published at the desired time and politicians are able to choose the tone of the messages themselves. Secondly, in order to get enough content for their stories journalists could now search for information on social media (Vergeer et al. 2013). Thirdly, via social media citizens are able to interact and have conversations with politicians, which was not possible with traditional media.
Some research is done on the effects of interactive communication and politics. In her research, Kruikemeier (2014) investigates the effects of interactivity and political personalization on the amount of preferential votes a candidate receives during an election. She found that candidates who made use of Twitter during their campaign received more preferential votes than the candidates who did not use this medium. This effect was even stronger for the candidates who used the medium interactively. Utz (2009) also found a positive effect of interactivity on the evaluation of a candidate. Interactivity on social media pages could increase political knowledge of the audience and result in a better evaluation of the candidate. Hwang (2013) investigated the effect of a politician’s Twitter use on this politician’s credibility and the attitude towards this politician in Korea. Based on his findings he concluded that a positive evaluation of a politician’s Twitter use leads to an increase of the perceived credibility of this politician.
Another important aspect of social media is the difference in social presence. ‘Social presence’ is the degree of salience of the other person during interpersonal communication when communicating via a certain medium (Gunawardena & Zittle, 1997). The extent to which a certain kind of contact is possible varies among different sorts of media (Short, Williams & Christie, 1976). When the social presence is high the influence that the communicator has on the audience is bigger than when social presence is low. Since the main goal of political campaigns is to get as much supporters as possible this is an important factor for the use of social media by politics. Two important aspects of the social presence theory are intimacy and immediacy. When communication is interpersonal and not mediated is perceived as more intimate and therefore social presence is perceived higher. The same applies to immediacy. When communication is synchronous and the communicators respond to each other immediately social presence is higher than with asynchronous
communication (Short, Williams & Christie, 1976). In their research Kruikemeier, Van Noort, Vliegenthart and De Vreese (2013) combine the aspects of interactivity and social presence in relation to political involvement. They found that online political communication that include both interactivity and high social presence had a stronger positive effect on citizens’ political involvement than media that do not allow a combination of these aspects. Social media are websites with a high social presence and a high form of interactivity. Therefore it is most effective for politics to use communication channels where interactive, interpersonal, and synchronous communication is possible.
While the increased use of social media by politicians is a general trend, the way politicians behave on these website varies tremendously (Crawford, 2009). Some politicians choose to keep the relationship with the audience formal and distanced, while other politicians choose to share information about their personal and private lives as well (Crawford, 2009; Vergeer et al., 2013). A general trend is that with the implementation of social media political campaigns seem to be more candidate-centered (Vergeer et al., 2013). This personalization of political campaign means that there is more focus on the candidates of the party than there is on the party as a whole (Mazzoleni & Schultz, 1999, Garzia, 2011). A result of this personalization is that messages are more effective and that the audience identifies more with the politicians (Kruikemeier et al., 2013).
Communication professionals and politics
During the last decade the landscape of media, politics, and political campaigns has changed tremendously (Scammell, 1999; Esser, Reinemann & Fann, 2000). Before the introduction of television political campaigns were labour-intensive. With the help of many volunteers politicians had to get out on the streets and organize meetings to get their message across. Nowadays campaigns are more capital-intensive. Media have become a central point of the campaigns and there is hardly any face-to-face communication with the voters (Scammell, 1999). Therewith campaigns are far more professionalized. Strömback (2007) defines professionalized campaigns as “being permanent, although with varying intensity; by the central campaign headquarters being able to coordinate the messages and the management of the campaign; and by using expertise in analyzing and reaching out to members, target groups and stakeholders, in analyzing its own and the competitors’ weaknesses and strengths and
making use of that knowledge, and in news management” (p. 56). Although these changes are mostly visible in the United States, they definitely occur in Europe too. The difference for Europe however, is that often the professionals are actually a part of the party already instead of coming from the outside (Esser et al., 2000).
The addition of communication professionals to campaign teams is not always seen as a positive development. Mainly this negative feeling is caused by a term that has developed over the last years: the ‘spin doctor’. A spin doctor is “someone who tries to influence the public opinion by putting a favorable bias on information when it is presented to the public or media” (Andrews, 2006, p. 23). People feel like politicians do not tell the whole truth, and therefore are not as honest as they are ought to be. This might suggest that when communication professionals administer political social media these negative feelings affect the evaluation of the political party or politician too. However, this does not mean that every communication professional is a spin doctor. In his research Gaber (2006) divides the work of communication professionals in ‘above-the-line’ and ‘below-the-line’. Above-the-line would be the work of the regular press officer, such as making official announcements, writing speeches, and scheduling interviews, while below-the-line would be creating a politician’s image and driving the news agenda. Although this division makes some sense, it could be a little naïve to think a clear distinction can be made. Spin doctors focus on the whole process of the political communication, so they also influence a politician’s speeches, the way official announcements are made and what a politician answers during interviews.
Communicating strategically does not always lead to the desired outcomes for the politician. By trying to take control over the news stories politicians try to manipulate journalists into positive coverage. However, journalists try to counter this manipulation and have started to cover politics differently. Journalists increasingly cover politics in a strategic frame, where there is more emphasis on the politician’s personalities, disagreements, and polls than on the important content of the election (Esser et al., 2001; De Vreese, 2004). This in turn has its effects on the public. When the public is exposed to these strategic frames they tend to become more cynical about politics (Cappella & Jamieson, 1996; De Vreese, 2004; De Vreese & Elenbaas, 2008). This spiral of cynicism eventually leads to a decrease of political participation and involvement (De Vreese, 2005). Although this is the main effect, there are some moderating factors. When someone is more knowledgeable about politics this effect is
stronger. Political self-efficacy influences this main effect also. When self-efficacy is low this effect is stronger than for people with high political self-efficacy (Cappella & Jamieson, 1996; De Vreese, 2004, De Vreese & Elenbaas, 2008).
This research
As discussed in the previous sections social media are integrated in political campaigns more and more. Social media could contribute to a more positive evaluation of a party or politician in multiple ways. Firstly, by adding these media to the campaign there is more room for interactive communication with the audience, which leads to a more positive evaluation of politicians. Secondly, the audience has the idea that the politician stands closer to the audience and is more approachable. A trend that came with the implementation of the social media is that the campaigns became even more candidate-centered than they already were. Therefore there audience expects to communicate with the politician itself. However, politicians are not always able to manage their account properly. Letting a communication professional take over this task could be a solution, was it not that the use of communication professionals in politics often causes a negative feeling among the public. It is expected that when a communication department starts taking over the administering of the social media account this would affect the evaluation of the politician and the messages negatively. This leads to the following hypotheses:
H1: When the communication department administers a political party’s or politician’s social media account, this party or politician will be evaluated lower than when the account is administered by the party or politician itself. H2: When the communication department administers a political party’s or politician’s social media account, the messages on this account will be evaluated lower than when this account is administered by the party or politician itself.
There are social media pages for both political parties and for politicians. Since campaigns have become so candidate-centered it is expected that the effect will be stronger for personal pages than for party pages. This leads to the following hypotheses:
H3: The type of social media page moderates the effect of the type of communicator on the evaluation of a party or politician. The effect that is expected in Hypothesis 1 will be stronger for a politician’s page than for the page of a political party.
H4: The type of social media page moderates the effect of the type of communicator on the evaluation of the social media messages. The effect that is expected in Hypothesis 2 will be stronger for a politician’s page than for the page of a political party.
Method Participants
Of all the 158 persons that started participating in this Internet experiment, a total of 124 participants have completed the questionnaire. Of these participants 29 percent is male and 71 percent is female. The mean age of the participants is 31 years old, with a minimum age of nineteen and a maximum age of 76 years old. Overall the participants are predominantly right politically oriented. The mean score on an eleven-point left (1) – right (11) scale is 6.20, with a standard deviation of 2.19.
Procedure
To answer the research question and test the hypotheses an Internet experiment is held. Participants are gathered via a snowballing sample. By clicking on a link the participant could fill out a survey. The questionnaire started with a screenshot of a Facebook page of either a politician or a political party. The participants have had a minute to view this page. After this, they were asked to evaluate the party or politician, depending on the condition, and the posted messages on this Facebook page. In order to find out if there are other factors influencing the main question the participants are asked about their political interest and involvement, their social media use and how they receive political information.
Manipulation
This Internet experiment has a 2 x 2 design. ‘Message communicator’ is the independent variable of this research, which is manipulated. The message
communicator is the person who administers the Facebook account, in this case the political party, politician, or the communication department of the party. In total there are four different groups. The participants are exposed to either a (1) Facebook page of a political party where a politician communicates the messages, (2) a Facebook page of a political party where a media professional communicates the messages on behalf of the party, (3) a Facebook page of a politician where the politician communicates the messages, and (4) a Facebook page of a politician where a media professional communicates the messages. On the pages where the communication department administers the account, the following sentence is added to the description of the page: “This account is administered by the communication department of the party Totaal Lokaal”. Thereby the posts on these pages are written in the third finite form, while on the other pages the messages are written in the first finite form. On the pages where the party or the politician was the communicator no extra sentence was added (see attachment 1). The participants are randomly assigned to one of the four conditions. Although several kinds of social media are popular, for this research it is chosen to only include the social medium ‘Facebook’. Facebook is interesting and accessible to the whole population and various types of content could be shared (Gil de Zúñiga, Jung & Valenzuela, 2012).
To improve the external validity of this research all the Facebook pages are from the same political party or politician. There is chosen for the local, unknown party Totaal Lokaal from the municipality of Súdwest-Fryslând. By choosing this party it is expected that none of the participants have prior judgements about this party and therefore there are no other factors influencing the results of this research other than manipulation. Therewith the name of the party Totaal Lokaal (Totally Local) has no connotations with either the left or the right of the political spectrum.
Manipulation check
To find out if the participants even recognized the manipulation two questions are asked. First it is asked if the participant knew who the communicator of the messages was. Secondly it is asked in which finite form the messages were written.
In this research there are four conditions: (1) party page, with no extra sentence and posts written in the first finite form, (2) party page, with the extra sentence and posts written in the third finite form, (3) politician page, without extra sentence and posts written in the first finite form, and (4) politician page, with the
extra sentence and posts written in the third finite form. Overall this difference was clear. Of all the participants who were in the first condition the majority (66 per cent) did not know who the communicator was. This is not surprising, since this was not emphasized. In this condition, 72 per cent of the participants knew the messages were written in the first finite form. In the second condition the results are less outspoken. Half of the participants knew it was the communication department posting messages on the party’s Facebook account and 43 per cent did not know or did not pay attention to this. Of all the participants in condition two, seventy per cent claims the posts were written in the first finite form. However, these messages were formulated in the third finite form. The results for the personal politician’s page are more accurate. Of all the participants who were in the third condition, 67 per cent thought the politician was the communicator himself, and 77 per cent knew the posts were written in the first finite form. Finally, of the participants in the fourth condition 82 per cent claims the communication department communicated the messages on the Facebook page on behalf of the politician and answered that the posts were written in the third finite form.
Table 1a. Manipulation check – Communicator.
Condition
1 2 3 4
Right answer 6.9% 50.0% 66.7% 81.4%
Wrong answer 27.6% 6.7% 13.3% 8.8%
Does not know 65.5% 43.3% 20.0% 8.8%
Table 1b. Manipulation check – Finite forms.
Condition
Right answer 72.4% 16.7% 76.7% 82.4%
Wrong answer 13.7% 76.7% 13.3% 5.8%
Does not know 13.8% 6.7% 10.0% 11.8%
Variables
Demographics. The general demographics ‘age’ (M = 30.56, SD = 13.10), ‘gender’ (29 per cent male), ‘education level’ (M = 4.9, SD = 1.89), and ‘political orientation’ (M = 6.20, SD = 2.19) were assessed.
Evaluation of political party/politician. The evaluation of a political party or politician is one of the two dependent variables in this research. This latent variable is measured on six dimensions: (1) sympathetic, (2) competence, (3) good leadership abilities, (4) intelligence, (5) trustworthiness, and (6) honesty (Utz, 2009). These six statements could be answered on a five-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’.
In order to use this latent variable in the analyses, one variable has to be created. First the variable for the evaluation of the party is created. A principal component analysis (PCA) is run to know if the six items measure the same component. There turned out to be two components. The first component includes the items ‘sympathetic’, ‘trustworthiness’, and ‘honesty’, and is therefore more focussed on the personal aspects (EV = 3.30, R2 = .47) of the party. The second component consists of the items ‘competence’, ‘good leadership abilities’, and ‘intelligence’, and is therefore more focussed on competency aspects (EV = 1.13, R2 = .16). To test the reliability of the scale of both components, a Cronbach’s alpha test is run. The reliability of the scale ‘Personal evaluation’ is good (α = .71, M = 3.56, SD = .46). This scale cannot be improved when items are deleted. Also the scale for the component ‘Competence evaluation’ is good (α = .82, M = 3.13, SD = .61), and cannot be improved by deleting items. Therefore two variables are created, namely ‘Personal evaluation of party’ and ‘Competence evaluation of party’.
The same analysis is done for the evaluation of the politician. The principal component analysis (PCA) shows that there are two components, again one that focuses on the personal aspects (EV = 3.78, R2 = .54), and one that focuses on the competency aspects (EV = 1.19, R2 = .17) of the politician. These components consist
of the same items as with the evaluation of the party. With use of the Cronbach’s alpha test the reliability of the two scales are tested. The reliability of the scale for the component ‘Personal evaluation’ is good (α = .84, M = 3.20, SD = .57), and cannot be improved by removing items. The same goes for the scale of the component ‘Competence evaluation’. The reliability of the scale is good (α = .80, M = 3.08, SD = .60), and cannot be improved by deleting items. For the evaluation of the politician two variables are created: ‘Personal evaluation of politician’ and ‘Competence evaluation of politician’.
Evaluation of social media messages. The second dependent variable is the ‘evaluation of social media messages’. Message assessment could be both cognitively and affectively (Slater & Rouner, 1996). This research is focussed on the affective message evaluation. The message evaluation measured on two different levels: (1) credibility of the messages and (2) trustworthiness of the message. The statements could be answered on a five-point Likert scale, ranging from ‘strongly disagree’ ‘strongly agree’.
Since this variable is a latent variable, a principal component analysis (PCA) has to be carried out for both the evaluation of the messages on the party’s page as for the messages on the politician’s page. Firstly, the PCA factor analysis for the messages on the party’s page shows that the two items measure one factor (EV = 1.59, R2 = .79). To test the scale’s reliability the Cronbach’s alpha test is run. The reliability of the scale is good (α = .73, M = 3.59, SD = .71), and cannot be improved. Secondly, the PCA is run for the messages on the politician’s page. This analysis shows that also for this variable the two items measure one component (EV = 1.70, R2 = .85). The Cronbach’s alpha test proves that the scale is good (α = .81, M = 3.40, SD = .72), and cannot be improved.
News media use for political information. News media use will be measured by combining nine items (Gil de Zúñiga et al. 2012). For nine media the participants are asked how often they consulted each medium in order to get information about politics. The questions could be answered on an eight-point Likert scale, ranging from ‘never’ to ‘every day’. The media included in this research are newspapers (M = 2.40, SD = 2.51), television (M = 3.80, SD = 2.36), current affairs programs (M = 2.86, SD = 2.16), radio (M = 2.43, SD = 2.65), news websites (M = 4.16, SD = 2.67), websites
of political parties (M = .33, SD = 1.12), social media (M = 3.15, SD = 1.12), and news apps on tablet or smartphone (M = 3.73, SD = 2.93).
Social media use. As stated in the theoretical discussion of this research social media are “web-based services that allow individuals to (1) construct a public or semi-public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections to those made by others within the system” (Boyd & Ellison, 2008, p. 211). ‘Social media use’ is the amount of time a person spends on social media. The measurement of this variable is based on the measurement of Gil de Zúñiga et al. (2012). Participants are asked how often they use social media in a regular week. These questions could be answered on an eight-point Likert scale, ranging from ‘never’ to ‘every day’. In this research the six most popular social media over the world are included, namely: Facebook, Twitter, Instagram, LinkedIn, Pinterest, and Tumblr. With the sample of this research Facebook is the most popular social medium (M = 5.95, SD = 2.20). Twitter (M = 1.67, SD = 2.48), Instagram (M = 2.48, SD = 3.07), and LinkedIn (M = 2.07, SD = 2.15) are used only two or three times a week. Pinterest (M = .41, SD = 1.21) and Tumblr (M = .28, SD = 1.15) are the least popular social media among the participants of this research.
Social media use for political information. Motivations for participants to use social to get political information will be measured based on Kaye and Johnson’s (2002) measurement of why people use the Internet for political information. Statements are given about reasons for using social media politically. These statements could be answered on a five-point Likert scale ranging from (1) ‘strongly disagree’ to (5) ‘strongly agree’.
Table 2. Social media use for political information.
M SD
To help me choose who to vote for. 2.33 1.27
To form an opinion about important issues. 2.48 1.25 To see what a candidate will do when he is elected. 2.35 1.17 To judge personal qualities of candidates. 2.23 1.17
Because the information is easy to obtain. 2.83 1.41 To find the political information I am looking for. 2.25 1.25 To keep up with the main issues of the day. 2.66 1.35
It is entertaining. 2.62 1.34
It is exciting. 2.11 1.04
To give me something to talk about with others. 2.27 1.13 To use as ammunition in arguments with others. 2.28 1.17
Results Explorative results
In order to find out if there are relations between the explorative variables and the dependent variables Pearson product-moment correlation coefficients are conducted. There are no significant relationships found between the demographic variables and the dependent variables ‘personal evaluation of party or politician’, ‘competence evaluation of party or politician’ and ‘evaluation of social media messages’ (see table 3).
When looking at the explorative variables ‘news media use for political information’, ‘social media use’, and ‘social media use for political information’, only ‘the use of websites of political parties for political information’ has a significant relationship with the dependent variables. There is a positive correlation between the
Table 3. Correlations between demographic variables and dependent variables.
Personal evaluation Competence evaluation Message evaluation
r p r p r p Age N = 124 -.109 .228 -.039 .669 -.093 .306 Gender N = 124 -.046 .610 .000 .996 -.099 .276 Education level N = 123 .080 .391 .068 .452 .049 .588 Political orientation N = 123 -.095 .296 -.041 .653 -.047 .606
use of websites of political parties and the personal evaluation of a party or politician, r = .342, n = 124, p = .000. There is also found a positive correlation between the use of these websites and the competence evaluation of a party or politician, r = .359, n = 124, p = .016. This indicates that there is a positive, moderate relationship between consulting websites of political parties for political information and the personal and competence evaluation of a party or politician. When a person consults political websites for political information, this person evaluates a political party or politician more positive, or vice versa. There is also found a positive correlation between consulting political websites for political information and the evaluation of the messages on a party’s or politician’s social media page, r = .286, n = 124, p = .001. There is a weak, positive relationship between the consulting of political parties’ websites and the evaluation of social media messages of parties and politicians. When a person consults political parties’ websites, the social media messages of a party or politician will be evaluated more positives, or vice versa.
Testing hypotheses
Evaluation of the party or politician. The main purpose of this research is to find out if parties and politicians are assessed differently when they are not administering their own social media accounts themselves, but if their communication professionals take over this task. To test the first hypothesis a one-way between subjects ANOVA was conducted. The effects of the ‘type of communicator’ on the ‘evaluation of the party or politician’ had to be compared, where for one condition the communicator is the party or politician itself and for the other condition the communicator is the communication department of the party. Because the variable ‘evaluation of the party or politician’ loaded on two components, two variables had to be created. Therefore the test had to be run twice, one time for the variable ‘Personal evaluation’ and another time for the variable ‘Competence evaluation’. There is not found a significant effect of the type of communicator on the personal evaluation of the party or politician [F(1,121) = 3.63, p = .06]. It cannot be concluded that the personal evaluation is significantly lower when a communication professional administers a social media account on behalf of a party or politician (M = 1.64, SD = .26) than when the party or politician does this itself (M = 1.72, SD = .26). However, the level of significance is only passed marginally. This insinuates that there might be an effect, but that is was not significant enough in this research. When looking at the
competence evaluation there is not found a significant main effect for the type of communicator on the competence evaluation of the party or politician either [F(1,121) = .01, p = .94]. Therefore it cannot be concluded that the competence evaluation of a political party or politician is significantly lower when a communication professional administers the social media account (M = 1.53, SD = .30) than when the party or politician does this itself (M = 1.55, SD = .27). Based on these finding the first hypothesis of this research has to be rejected.
It was expected that the type of social media page would moderate the effect of type of communicator on the evaluation of the party or politician. To test this hypothesis a two-way univariate ANOVA is conducted, where the moderating effect of the type of social media page on the effect of type of communicator on the personal and competence evaluation of the political party or politician have to be compared. There is not found a significant moderating effect of type of social media page on the effect of communicator on the personal evaluation on the political party or politician [F(1,121) = .60, p = .44]. However, there is found a significant main effect of the type of social media page on personal evaluation of the party or politician [F(1,118) = 15.3, p < .00]. The personal evaluation of political parties or politicians is more positive for party pages (M = 1.76, SD = .23) than it is for politician’s pages (M = 1.58, SD = .26). The same two-way univariate ANOVA is run to test if the type of page moderates the effect of communicator on social media pages on the competence evaluation of the party or politician. No significant moderating effect is found [F(1,116) = .03, p = .86]. Based on these findings this third hypothesis has to be rejected.
Evaluation of social media messages. Another important question of this research is that if a communication department administers a political party’s or politician’s social media account the posted messages will be evaluated differently than when a party or politician does this itself. It is expected that when the communication department takes over this task the messages will be evaluated less positive than when the party or politician posts the messages itself. To test this hypothesis a one-way univariate ANOVA is conducted, where the ‘type of communicator’ and the ‘evaluation of the social media messages’ had to be compared. There is not found a significant effect of the type of communicator on the evaluation of the social media messages [F(1,121) = 1.00, p = .32]. Therefore it cannot be concluded that the
evaluation of social media messages is significantly lower when the communication department takes over the task of administering the social media account (M = 1.71, SD = .38) than when the party or politician administers the account itself (M = 1.77, SD = .33) and hypotheses two has to be rejected.
Furthermore it was expected that the type of social media page would influence the effect of the type of communicator on the evaluation of the social media messages. To test this hypothesis a two-way univariate ANOVA is conducted. There is not found a significant moderating effect of the type of social media page on the effect of the type of communicator on the evaluation of the social media messages [F(1,119) = .83, p = .36] and the fourth hypothesis has to be rejected also.
Conclusion and discussion
The main purpose of this research was to find out what effect the deployment of communication professionals on social media of political parties and politicians have on the evaluation of these parties and politicians and the messages that are posted on their social media page. Social media have become a very important media in today’s society. While for the public the main purpose of the use of social media is to stay connected with their current social network (Utz, 2009), many politicians have integrated these media in their campaigns too (Crawford, 2009; Karlsen, 2010). As explained in the literature, implementing social media in a political campaign could be beneficial for politicians in several ways. Firstly, via social media politics can bypass the mass media, and therefore choose their own frame and own time to get their message across (Bakker, 2013). Secondly, it is possible to interact with the audience (Kruikemeier, 2014). Via social media politicians have the ability to communicate with the public, which was not so easy before. Thirdly, social media are media where social presence is high and therefore the communicator has more influence on the audience that with media with low social presence (Short et al., 1976). Several studies have proven that the combination of these aspects lead to a better evaluation of a politician (Utz, 2009; Kruikemeier et al. 2013; Hwang, 2013, Kruikemeier, 2014). A fourth benefit of implementing social media in the campaign is that the audience can identify more with the politicians. However, these benefits only occur when the social media are used properly. Administering a social media account takes time and effort. Also, when information is on the Internet it will not go away. Therefore it is essential that the messages that are formulated properly and do not
harm a politicians image. Normally communication professionals make sure messages are well formulated and publish at the right moment. However, since candidates are the central focus of their social media pages the audience expects that the politician administer their own account (Vergeer et al., 2013; Kruikemeier et al., 2013). Therewith the use of communication professionals in politics has a negative connotation, because this suggests that politicians only communicate the information that is beneficial to them instead of telling the whole truth (Andrews, 2006).
To find out if it would lead to a more negative evaluation of the political party or politician and the social media message an Internet experiment is held where the communicator was manipulated. Based on their condition participant were exposed to either a social media page of (1) a political party, (2) a political party that was administered by the communication department, (3) a politician, or (4) a politician that was administered by the communication department. There are no significant differences found between the difference in a communicator on social media pages and the evaluation of the political parties or politicians and the messages. Therefore the hypotheses cannot be accepted. However, it cannot be concluded that deploying a communication professional on political social media has no effect on the evaluation of the party, politician, or messages either. Therefore answering the research question is not possible.
There are various factors that could have led to the insignificant results. Firstly, the manipulation might have been not explicit enough. The manipulation check showed that participants in the second condition, where the communicator was the communication department and messages were written in the third finite form, only half of the participants knew that the communication department was administering the account and more than three-quarters of this group thought the messages were written in the first finite form. Secondly, to improve the external validity of this research there was chosen to expose the participant to a social media page of an unknown party or politician. By doing so there would be no prejudgments influencing the evaluation. However, now participants were asked to evaluation a political party or politician based on only a screenshot of a social media pages, which might not be enough information to judge a political party or politician on. Thirdly, interactivity with the politician, an aspect that distinguishes social media from traditional media, was not possible in this Internet experiment. Participants were exposed to a social media page, but there was no opportunity to actually interact with
the politicians. Therefore this aspect that mostly affects the evaluation of politicians positively was absent in this research. When the influence of the different communicator on the personal evaluation of the party or politician was analyzed, there was not found a significant effect. However, the significance level was only marginally passed. This implicates that there might be an effect, but that is was not strong enough in this research.
These implications taken together leads to the conclusion that although there are not found any significant results it might be useful to do further research on this topic. Suggested research would be a longitudinal study, where participants actually follow specific parties and politicians on social media. Then there is the opportunity to include interactivity and participants have more information to base their evaluations on.
References
Andrews, L. (2006). Spin: from tactic to tabloid. Journal of Public Affairs, 6(1), 31-45.
Bakker, T. P. (2013). Citizens as political participants. The myth of the active online audience? (Unpublished doctoral dissertation). University of Amsterdam. Chapter 1: Introduction.
Boyd, D. M., & Ellison, N. B. (2007). Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13, 210–230.
Cappella, J. N., & Jamieson, K. H. (1996). News frames, political cynicism, and media cynicism. The Annals of the American Academy of Political and Social Science, 71-84.
Crawford, K. (2009). Following you: Disciplines of listening in social media.
De Vreese, C. (2004). The effects of strategic news on political cynicism, issue evaluations, and policy support: A two-wave experiment. Mass Communication & Society, 7(2), 191-214.
De Vreese, C. H., & Elenbaas, M. (2008). Media in the game of politics: Effects of strategic metacoverage on political cynicism. The International Journal of Press/Politics, 13(3), 285-309.
Drezner, D., & Farrell, H. (2007). The power and politics of blogs. Public Choice, 134(1-2), 15–30.
Esser, F., Reinemann, C., & Fan, D. (2001). Spin Doctors in the United States, Great Britain, and Germany Metacommunication about Media Manipulation. The Harvard International Journal of Press/Politics, 6(1), 16-45.
Garzia, D. (2011). The personalization of politics in Western democracies: Causes and consequences on leader–follower relationships. The Leadership
Quarterly, 22(4), 697-709.
Gil de Zúñiga, H., Jung, N., & Valenzuela, S. (2012). Social media use for news and individuals’ social capital, civic engagement and political participation. Journal of Computer- Mediated Communication, 17(3), 319 – 336.
Gunawardena, C. N., & Zittle, F. J. (1997). Social presence as a predictor of satisfaction within a computer‐mediated conferencing environment. American
journal of distance education, 11(3), 8 – 26.
Ha, L., & James, E. L. (1998). Interactivity reexamined: A baseline analysis of early business web sites. Journal of Broadcasting & Electronic Media, 42(4), 457 -474.
toward Politicians. Journal of Public Relations Research, 25(3), 246-258.
Kaplan, A. M. & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of Social Media. Business Horizons, 53, 59 – 68.
Karlsen, R. (2010). Does new media technology drive election campaign change? Information Polity, 15(3), 215 – 225.
Katz, E., Blumler, J. G., & Gurevitch, M. (1973). Uses and gratifications research. Public opinion quarterly, 509-523.
Kaye, B. K., & Johnson, T. J. (2002). Online and in the know: Uses and gratifications of the Web for political information. Journal of Broadcasting & Electronic
Media, 46(1), 54-71.
Kruikemeier, S. (2014). How political candidates use Twitter and the impact on votes. Computers in Human Behavior, 34, 131-139.
Kruikemeier, S., van Noort, G., Vliegenthart, R., & de Vreese, C. H. (2013). Getting closer: The effects of personalized and interactive online political communication. European Journal of Communication, 28(1), 53-66.
Mangold, W. G., & Faulds, D. J. (2009). Social media: The new hybrid element of the promotion mix. Business horizons, 52(4), 357-365.
Mazzoleni, G., & Schulz, W. (1999). " Mediatization" of politics: A challenge for democracy?. Political Communication, 16(3), 247-261.
Quan-Haase, A., & Young, A. L. (2010). Uses and gratifications of social media: A comparison of Facebook and instant messaging. Bulletin of Science,
Technology & Society, 30(5), 350-361.
Scammell, M. (1999). Political marketing: lessons for political science. Political
Shao, G. (2009). Understanding the appeal of user-generated media: a uses and gratification perspective. Internet Research, 19(1), 7-25.
Slater, M. D., & Rouner, D. (1996). How message evaluation and source attributes may influence credibility assessment and belief change. Journalism & Mass
Communication Quarterly, 73(4), 974-991.
Strömbäck, J. (2007). Political marketing and professionalized campaigning: A conceptual analysis. Journal of Political Marketing, 6(2-3), 49-67.
Utz, S. (2009). The (potential) benefits of campaigning via social network sites. Journal of Computer-Mediated Communication, 14(2), 221–243.
Van Aelst, P., Shehata, A., & Dalen, A. V. (2010). Members of Parliament: Equal competitors for media attention? An analysis of personal contacts between MPs and political journalists in five European countries. Political Communication, 27(3), 310-325.
Vergeer, M., Hermans, L., & Sams, S. (2013). Online social networks and micro-blogging in political campaigning The exploration of a new campaign tool and a new campaign style. Party Politics, 19(3), 477-501.
Wang, Z., Tchernev, J. M., & Solloway, T. (2012). A dynamic longitudinal examination of social media use, needs, and gratifications among college students. Computers in Human Behavior, 28(5), 1829-1839.
Attachment 1. Manipulation Condition 1:
Condition 3: