Getting closer: The effects of personalized and interactive online political communication

Abstract

Political parties and politicians increasingly use the possibilities of the Internet to communicate interactively with citizens and vice versa. The Internet also offers opportunities for individual politicians to profile themselves. These developments are often said to bring politics closer to citizens, increasing their political engagement in politics. Empirical evidence for such claims is, however, scarce. In a scenario experiment and a laboratory experiment using real-world websites, the authors examine whether more personalized online communication (a focus on individual politicians) and the use of interactive features increase political involvement among citizens. The results from both studies demonstrate that both highly interactive and personalized online communication do increase citizens’ political involvement. Moreover, it was also found that political personalization positively moderates the effect of interactivity on political involvement, meaning that the effects of interactivity are even stronger in a personalized setting.

Keywords

Experiment interactivity Internet personalization political involvement

Political parties and elected representatives increasingly use the possibilities of the Internet to communicate interactively with citizens and party members about their plans, points of view and daily business (e.g. Druckman et al., 2007; Kenski and Stroud, 2006). For citizens, the Internet offers easy access to political information, providing all kinds of opportunities to participate in political debates.

The Internet not only enables interactive communication between political parties and citizens, but social media such as Facebook and Twitter, also offer opportunities for individual politicians to profile themselves. According to Van Santen and Van Zoonen (2010), the rapidly growing number of politicians using new online communication instruments could be seen as a form of ‘personalization’ – the shift of attention from political parties to politicians, especially since these new media technologies are designed to facilitate direct communication between politicians, not parties, and citizens.

It has been claimed that online political communication (that facilitates interactive and personalized communication), may increase citizens’ political engagement, by bringing politics closer to citizens (e.g. De Vreese, 2007; Tolbert and McNeal, 2003). There is, however, little empirical evidence for these claims (Boulianne, 2009). Hence, it remains unclear whether personalization and the use of specific interactive features should be part of an effective digital political communication strategy. Our study aims to tackle this uncertainty and contributes to the existing literature in three ways. First, most of the work on online political communication has been devoted to the effect of political Internet use in general (e.g. Kenski and Stroud, 2006; Shah et al., 2005) and not to the effects of specific characteristics of new media. Our study instead investigates the effects of two specific characteristics: interactive and personalized communication. Second, to date, the literature on political personalization has rarely addressed the consequences of personalization, especially in online communication. Third, previous studies have not yet examined the combined impact of personalization and interactivity on citizens’ political involvement. The central question in this article is: To what extent do levels of political personalization and interactivity of online communication increase political involvement among citizens?

To answer this question, we conducted two experiments. In the first experiment we use data from a survey-embedded scenario experiment, which we distributed to a representative sample of the Dutch population. The second experiment replicates and extends the findings in the first experiment, by using a laboratory experiment with real-world websites, different research population and extended measures.

Interactivity and personalization in online political communication

The notion of interactivity is widely used in various disciplines, including political communication. It suffers, however, from conceptual confusion and contradictions. The different understandings have in common that they assume two-way communication, which is also how we will define interactivity in this study. Research about the consequences of interactivity mostly finds positive effects on individual level political variables, such as more positive evaluations of candidates (Sundar et al., 2003) and a heightened sense of political efficacy (Tedesco, 2007). In line with these arguments, we hypothesize that high interactive online political communication has a positive effect on citizens’ political involvement (H1).

Personalization suffers from similarly diverging conceptualizations (Van Santen and Van Zoonen, 2010), with the common denominator being that personalization involves a shift in public attention from political institutions and parties, to politicians and their individual competences, private lives and emotional reflections (sometimes referred to as political privatization; Adam and Maier, 2010; Rahat and Sheafer, 2007; Van Santen and Van Zoonen, 2010). Personalization might take place especially in new media channels: the specific features used in new and social media ‘are designed to facilitate a direct link between sender (politician) and receiver (citizen), and vice versa’ (Van Santen and Van Zoonen, 2010: 65), which consequently puts individual politicians more to the forefront. Only a few studies have addressed the consequences of political personalization and found positive effects, including increased effectiveness of messages (for an overview, see Brettschneider, 2008), identification with politicians (Langer, 2007) and possibly electoral turnout (Kleinnijenhuis et al., 2001; Rosenberg et al., 1986). By taking these outcomes as a starting point, we propose that personalized online political communication has a positive effect on citizens’ political involvement (H2).

Finally, if interactivity and personalization are combined in online political communication, it is likely that possible positive effects may even be stronger. This expectation is based on the Social Presence Theory (Short et al., 1976; Tanis, 2003). Social presence increases when a medium (or its features) resembles interpersonal communication (Fortin and Dholakia, 2005; Tanis, 2003). Since interactive and personalized communication closely resemble interpersonal communication (because it enables two-way communication and the communication exchange becomes ‘personal’), the perception of ‘social presence’ of the politician will be even stronger, than in the solely interactive or only personalized case (Short et al., 1976; Tanis, 2003). Hence, we assume that a positive effect of high interactive online political communication on citizens’ political involvement is stronger when the online political communication is also personalized, and vice versa (H3).

To test our three hypotheses, we conducted a real-life, scenario-based survey-embedded experiment (Study 1) and a laboratory experiment with real-world websites as stimuli (Study 2).

Study 1: Design, measures and outcomes

In the first experiment, we designed a survey in which we offered the respondents a scenario about a fictitious political website, which we manipulated in terms of content and level of interactivity. We created four scenarios on the basis of high–low interactivity and high–low personalization (see Figure A1 in Appendix). In the high interactive condition, the scenario described a website of a party that actually exists (the social-liberal party D66, a small size, centre political party in the Netherlands), and which contained features that enable two-way communication and facilitate control over the communication process, whereas in the low interactivity condition features were described that only allow for one-way communication, and do not allow for control (based upon Liu and Shrum, 2002; Voorveld et al., 2010). In the high personalization scenario the focus of the website was on individual competence and private lives, whereas the low personalization scenario offered information about the organization and history of the political party (based upon Van Santen and Van Zoonen, 2010). Two pre-tests showed that personalization and interactivity were effectively manipulated.

Participants from a representative sample of the Dutch population were invited by email in March 2012. The response rate was 68.4% (calculated according to AAPOR guidelines; AAPOR, 2011), with 718 people responding (52.8% female, mean age 49). After outliers were removed, 650 respondents remained in the sample. After filling out a questionnaire, they were directed online to one of the four scenarios. Next, participants were asked to vividly imagine a situation in which they are searching for political information online, because of an upcoming election.

After reading the scenario, respondents answered questions about political involvement, particularly about having the opportunity to come in contact with politics or politicians and their feelings of being close to politics or politicians. ‘Contact’ was measured using two items (i.e. ‘this website offers opportunities for a dialogue’ and ‘this website gives me the opportunity to come into contact with politicians’). Answers were coded on a seven-point scale (1 = totally disagree, 7 = totally agree; M = 3.79, SD = 1.33). ‘Closeness’ was also measured by two items (i.e. ‘this website gives me the feeling that I am closer to politics’ and ‘this website gives me the feeling that politicians are more involved with their electorate’). Answers were coded on a seven-point scale (1 = totally disagree, 7 = totally agree; M = 3.46, SD = 1.34). A confirmatory factor analysis (CFA) with varimax rotation showed that the validity and reliability of our scales were good (details are available upon request from the authors). Last, we included the control variable likelihood of voting for D66 in our analyses to test for possible confounds. This was measured by asking participants: ‘How likely is it that you would vote for D66?’ Answers were recorded on an 11-point scale (1 = I would never vote for this party, 11 = I would certainly vote for this party; M = 5.34, SD = 3.29).

The results of the multiple analysis of covariance (MANCOVA) show that there are effects of interactivity on both the perception of how easy it is to come into contact with politics, and on feelings of closeness to politics. The difference between high and low was significant for both indicators. People in the high interactivity condition scored on average 0.52 points higher on the scale that indicates how easy they consider it to be in contact with politics and 0.18 points higher for their feelings of closeness to the politics. Thus, participants in the high interactive condition felt that they had more opportunities to come in contact with, and felt closer to politics than participants in the low interactive condition. These results support our first hypothesis.

In terms of the effect of personalization, the results demonstrate that participants in the personalized condition felt that they had more opportunities to come into contact with politics (0.40 points higher on the scale) than participants in the non-personalized condition. However, the results did not reveal an effect of personalization on the sense of closeness to politics and politicians. Thus, our second hypothesis was only partly supported.

With regard to the interaction effects of interactivity and personalization on contact and feelings of closeness, we found a significant interaction between interactivity and personalization on the contact scale; more specifically, the effect of interactivity was significant in the personalized scenario, but only marginally significant in the non-personalized condition. Results on the closeness scale show similar results: the effect of interactivity was significant in the personalized condition but not in the non-personalized condition. This means that the effects of interactivity on political engagement only occur in personalized political communication. These results confirm our third hypothesis (see Table A1 in Appendix).

Study 2: Design, measures and outcomes

In the second study we tested the same hypotheses, by conducting a laboratory experiment with real-world websites. Participants were 262 undergraduate students (74.4% female) with a mean age of 21 (SD = 2.61). We manipulated the level of personalization and interactivity in the same way as in our first study, using the actual website of the social-liberal party D66 as a basis (a pre-test showed that personalization and interactivity were effectively manipulated). This resulted again in a 2 (low vs high interactivity) × 2 (non-personalized vs personalized) between subjects design (see Table A3 and Figure A2 in Appendix). Participants were randomly assigned to one of the four websites and were asked to evaluate the usability of the site, and to answer questions about political involvement with the party and the politician. We excluded participants who were familiar with the real-world website of D66 (7.7%) to reduce method bias.

We measured political involvement by four dependent variables (i.e. contact, closeness, arousal of political interest and intention to revisit the website). Contact was this time measured using five items (i.e. ‘this website gives me the opportunity to come into contact with politicians’, ‘invites people for a conversation’, ‘shows that politicians are open to ideas from citizens’, ‘gives politics the opportunity to react to ideas from citizens’ and ‘shows that politicians are willing to listen to me’). Answers were coded on a seven-point scale (1 = totally disagree, 7 = totally agree; M = 3.00, SD = 1.40). Closeness was measured using four items (i.e. ‘this website reduces the distance between me and politics’, ‘gives me the feeling that I am closer to politics’, ‘gives me the feeling that politicians are more involved with their electorate’ and ‘connects people’). Answers were coded on a seven-point scale (1 = totally disagree, 7 = totally agree; M = 3.42, SD = 1.25). Here too, a confirmatory factor analysis with varimax rotation revealed that these two scales were reliable and valid (details are available upon request from the authors). Arousal of political interest was measured using one item (i.e. ‘this website arouses my interest in politics’). Answers were coded on a seven-point scale (1 = totally disagree, 7 = totally agree; M = 2.87, SD = 1.31). Intention to revisit website was also measured using one item (i.e. ‘I have the intention to revisit the website’). Answers were coded on a seven-point scale (1 = totally disagree, 7 = totally agree; M = 2.04, SD = 1.21). We controlled for three variables that we expected to interfere with the main effects of interactivity and personalization that we are interested in: likelihood of voting for D66, affective and cognitive involvement with the website.

The results of our MANCOVA show that there is a significant effect of interactivity on political involvement, meaning that participants in the high interactive condition felt that they had more opportunities to come into contact with (on average 2.01 points higher), felt closer to (1.12 points higher) and had the feeling that the website aroused their interest in politics (0.57 points higher). However, the results revealed no main effects of interactivity on the intention to revisit the website. Overall, our first hypothesis is largely supported.

In terms of the effect of personalization, the results revealed that participants in the personalized condition felt that they had more opportunities to come into contact with politics than participants in the non-personalized condition (0.48 points difference). However, the results did not show an effect on closeness, the arousal of political interest, or intention to revisit the website. Thus, our second hypothesis was only marginally supported.

We also examined the interaction effects between personalization, interactivity and the dependent variables (see Table A2 in Appendix). The ANCOVA showed a significant interaction effect between interactivity and personalization on the various outcome variables. More specifically, the results reveal that the effects of interactivity are only significant in the personalized condition for political interest arousal, and intention to revisit the site. For contact and perceived closeness, the effects of interactivity are significant for both the personalized and non-personalized conditions, but in both instances the effect was significantly larger in the personalized condition. Figures 1 –4 provide a graphic representation of the interaction effects. Overall, these results offer support for our third hypothesis.

Figure 1.

Scores on the contact scale as a function of personalization and interactivity.

Figure 2.

Scores on the closeness scale as a function of personalization and interactivity.

Figure 3.

Scores on the arousal of political interest scale as a function of personalization and interactivity.

Figure 4.

Scores on the intention to revisit the website as a function of personalization and interactivity.

Discussion

In this article, we studied the effectiveness of personalization and interactivity in online political communication. We examined whether a focus on an individual politician (vs a party) and the use of interactive features increase political involvement among citizens. As expected, both studies show that personalization and interactivity have a positive effect on citizens’ political involvement. Citizens who visit a website which is more focused on an individual politician or which contains more interactive features, feel more politically involved than citizens who visit a website focused on a political party or which did not contain interactive features. Those results are particularly consistent with earlier studies which looked at the effects of interactivity (Sundar et al., 2003; Tedesco, 2007) and demonstrated positive and desirable effects of higher levels of interactivity. Furthermore, in accordance with our expectation based on the Social Presence Theory (Short et al., 1976; Tanis, 2003), we also found that the combined effect of personalized, interactive online political communication has an even stronger positive effect on citizens’ involvement than when taken separately. More precisely, we observed that interactive, personalized online communication has a positive effect on citizens’ feelings of having the opportunity to come into contact with politics, and citizens’ feelings of closeness to politics. This implies that it is effective to combine these two new media strategies. Of course, not all citizens will be equally influenced. Part of the electorate is likely to be suspicious towards interactivity and personalization and will consider them marketing tools to convince people, rather than genuine attempts to strengthen the ties between politics and voters, which may decrease the persuasive power (Friestad and Wright, 1994).

In Study 2, we additionally found that individuals who report that an interactive, personalized website arouses their interest in politics and individuals who visited the interactive, personalized website said more often that they would revisit this website, compared to citizens who did not visit this website. It seems that besides feelings of political involvement, personalized and interactive online communication may also exert positive effects on actual interest in politics and behaviour, which is also supported by previous evidence (Thorson and Rodgers, 2006). However, it is important to note that, although personalization and interactivity arouse interest and behaviour, the overall levels of ‘political interest caused by the website’ and ‘intention to revisit the website’ were rather low. This might indicate that personalized and interactive communication strategies are not sufficient to engender high levels of political interest and possible behaviour. Perhaps, citizens’ trait characteristics, such as efficacy and civic skills (Verba et al., 1995) are necessary for online political communication to have a strong positive impact on political interest and behaviour. Future research needs to investigate whether the effects of personalized and interactive communication are more pronounced for more politically sophisticated citizens.

Interesting for future research is also the fact that the results of the scenario study are similar to the ones found in the laboratory experiment. It could be argued that it is difficult for participants to vividly imagine an ‘interactive’ website on the basis of a description of a website. Especially since a description does not contain actual interactivity. Nonetheless, on the basis of our results, participants seem to be able to clearly visualize a website in their mind without actually visiting one. Therefore, the scenario design seem to be a valid method for future research that studies the effects of online communication.

To sum up, the findings presented in this article contribute to theory about the effects of political Internet use, by demonstrating that the characteristics of online communication can contribute to citizens’ political involvement. An interesting finding, since scholars who study the consequences of Internet use for political purposes disagree, and tend to be more pessimistic about the impact of Internet use on citizens’ political involvement (for an overview of the literature, see Boulianne, 2009). Our study shows a more positive view. It seems that the characteristics of new media are important elements that might eventually explain the positive consequences of political Internet use. Since political parties and their representatives are increasingly using more interactive features online and making their communication more individual (e.g. by using Twitter), political involvement among citizens could increase, which in the end will foster democracy. From this viewpoint, the characteristics of online media become not only more important when studying the effects of political Internet use, but also seem an important condition for the effect Internet may have on democracy.

Footnotes

Appendix

Table A3.

Personalization and interactivity in the manipulated websites in study 2.

Concept	Non-personalized, low interactive	Personalized, low interactive	Non-personalized, high interactive	Personalized, high interactive
Interactivity (hyperlinks)	0	0	11	11
Personalization (logo vs photo)	Logo D66	Photo AP	Logo D66	Photo AP
Personalization/ interactivity (background information)	Party information on homepage	Biography AP on homepage	Party information via hyperlink	Biography AP via hyperlink
Personalization/ interactivity (3 news items)	News items on homepage; D66 centre of focus	News items on homepage; AP centre of focus	News items, with hyperlink; D66 centre of focus	News items, with hyperlink; AP centre of focus
Personalization/ interactivity (commenting on news items)	–	–	Comments D66 and unknown visitors	Comments AP and unknown visitors
Interactivity (sharing news items on SNS)	–	–	+	+
Interactivity (emailing news items to others)	–	–	+	+
Interactivity (RSS feed)	–	–	+	+
Interactivity (contact)	–	–	+	+
Interactivity (participation in the campaign)	–	–	+	+
Interactivity (Twitter feeds displayed in website)	–	–	Twitter feeds D66	Twitter feeds AP
Interactivity (links to SNS and other websites)	–	–	+	+
Interactivity (link blog)	–	–	Link blog D66	Link blog AP

AP = Alexander Pechtold (party leader D66); not present = –; present = +

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

AAPOR (2011) Standard definitions: Final dispositions of case codes and outcome rates for surveys. 2011. Available at: www.aapor.org/AM/Template.cfm?Section=Standard_Definitions2&Template=/CM/ContentDisplay.cfm&ContentID=3156.

Adam

Maier

(2010) Personalization of politics: A critical review and agenda for research. In: Salmon

(ed.) Communication Yearbook 34. London: Routledge, pp. 213–257.

Boulianne

(2009) Does Internet use affect engagement? A meta-analysis of research. Political Communication 26(2): 193–211.

Brettschneider

(2008) Personalisation of campaigning. In: Donsbach

(ed.) The International Encyclopedia of Communication. Oxford: Blackwell.

De Vreese

(2007) Digital renaissance: Young consumer and citizen? The Annals of the American Academy of Political and Social Science 611(1): 207–216.

Druckman

Kifer

Parkin

(2007) The technological development of congressional candidate web sites. Social Science Computer Review 25(4): 425–442.

Fortin

Dholakia

(2005) Interactivity and vividness effects on social presence and involvement with a web-based advertisement. Journal of Business Research 58(3): 387–396.

Friestad

Wright

(1994) The persuasion knowledge model: How people cope with persuasion attempts. Journal of Consumer Research 21(1): 1–31.

Kenski

Stroud

(2006) Connections between Internet use and political efficacy, know-ledge, and participation. Journal of Broadcasting and Electronic Media 50(2): 173–192.

10.

Kleinnijenhuis

Maurer

Kepplinger

Oegema

(2001) Issues and personalities in German and Dutch television news. European Journal of Communication 16(3): 337–359.

11.

Langer

(2007) A historical exploration of the personalisation of politics in the print media: The British prime ministers (1945–1999). Parliamentary Affairs 60(3): 371–387.

12.

Liu

Shrum

(2002) What is interactivity and is it always such a good thing? Implications of definition, person, and situation for the influence of interactivity on advertising effectiveness. Journal of Advertising 31(4): 53–64.

13.

Rahat

Sheafer

(2007) The personalization (s) of politics: Israel, 1949–2003. Political Communication 24(1): 65–80.

14.

Rosenberg

Bohan

McCafferty

Harris

(1986) The image and the vote: The effect of candidate presentation on voter preference. American Journal of Political Science 1: 108–127.

15.

Shah

Cho

Eveland

Kwak

(2005) Information and expression in a digital age. Communication Research 32(5): 531–565.

16.

Short

Williams

Christie

(1976) The Social Psychology of Telecommunications. Chichester: Wiley.

17.

Sundar

Kalyanaraman

Brown

(2003) Explicating web site interactivity. Communication Research 30(1): 30–59.

18.

Tanis

(2003) Cues to identity in CMC: The impact on person perception and subsequent interaction outcomes. University of Amsterdam.

19.

Tedesco

(2007) Examining Internet interactivity effects on young adult political information efficacy. American Behavioral Scientist 50(9): 1183–1194.

20.

Thorson

Rodgers

(2006) Relationships between blogs as eWOM and interactivity, perceived interactivity, and parasocial interaction. Journal of Interactive Advertising 6(2): 39–50.

21.

Tolbert

McNeal

(2003) Unraveling the effects of the Internet on political participation? Political Research Quarterly 56(2): 175–185.

22.

Van Santen

Van Zoonen

(2010) The personal in political television biographies. Biography 33(1): 46–67.

23.

Verba

Schlozman

Brady

(1995) Voice and Equality: Civic Voluntarism in American Politics. Cambridge, MA: Harvard University Press.

24.

Voorveld

Neijens

Smit

(2010) Exploring the link between objectively and subjectively assessed interactivity on global brand websites. In: 60th Annual ICA Conference, Singapore.