Online election campaigning: exploring supply and demand during the France 2012

presidential election

Darren G. Lilleker (Bournemouth University, UK)

Karolina Koc-Michalska (Science-Po, Paris, France)

Abstract

Studies of online campaigning tend to focus on the supply side: the way political parties

communicate and campaign using the Internet. This chapter explores the online presences of

the main candidates and their parties who stood in the 2012 French presidential election. Our

research focuses not only on the supply side but also explores demand, utilising data from the

Mediapolis survey to ascertain what citizens search for online and in particular what citizens

seeking help with their voter decisions seek online. Our data shows that citizens are provided

with a rich online experience during election campaigns. Information is presented in

engaging ways and candidates attempt to mobilise their supporters and offer various

opportunities to interact with the campaign and other website visitors. Interaction is

augmented in particular by the use of social networking sites. Citizens, however, appear to

mostly go online to find detailed information on the policies and programmes of the

candidates. There appears little call for engaging communication, interactive opportunities or

details on the personal lives or personalities of the candidates. Our data may therefore suggest

that information may need to be packaged for accessibility, presented in a way that allows

voters to make up their own minds rather than following the norms of corporate sales

campaign websites.

Introduction

A plethora of pseudo-scientific sociological, business or marketing texts offer the impression

that the Internet has the capacity to revolutionise all aspects of modern life. The mundane

behaviours relating to buying goods, how we communicate, how we build a social circle, and

how we influence the world around us have been revolutionised (Baym, 2010; Press &

Williams, 2010). The evolution of technological development which ushered in Web 2.0

platforms and tools, in particular, lead to talk of big conversations taking place at a global

level; superseding national institutions as well as nations themselves. There is no doubt that

the Internet has had a clear and tangible impact on the lives of many, though not all, who live

in modern societies. However reality does not always live up to the hype. Social media

platforms, such as Twitter, Facebook and YouTube, have become part of the toolkit of the

strategic communicator. Social media facilitate a more immediate, more widely connected,

communication environment that can be used for any purpose the individual desires. People

can create first-hand accounts from within Syria, or amusing videos of cats; the audience

equally can choose which they want to watch during their leisure time. What the various

works on the potential and actual impact of the Internet tell us is that a lot of potential for

change exists, but that both creators and audiences seem reluctant to explore this potential.

The dichotomy between the potential and the actual is nowhere more clear than within the

context of electoral politics and the extent the Internet will lead to a more participatory

democratic culture.

This chapter focuses on one election contest: the French presidential election of 2012. We set

the scene by outlining the potential that the Internet might play in terms of usage by political

parties and candidates and the reality of the role played by online communication as shown

by a range of research projects over the last decade or so. The data we go on to present is

unique. We are able to describe and explain the way the Internet was used during the election

contest, while also outlining how the French voters used the Internet. We thus provide a

picture of the campaigns of the candidates and their supporting parties, judging the extent that

it was both persuasive and participatory. To measure the demand on such a political content

we provide data on the forms of online political engagement and on the characteristics of

those who are most politically active online. .

Political communication and elections online

With the Internet becoming a publicly accessible medium emerged a range of optimistic

accounts of the impact upon politics. The cyber-optimist position argued that existing

hierarchies of power and influence would be flattened, any individual would have an equal

share of voice independent of their status and politics would develop an inclusive and

participatory character (Rheingold 1993; Negroponte 1995; Dertouzos 1997). The arguments

surrounding the potential uses and impacts of the Internet are summarised well by Pippa

Norris (2003: 24):

first wave advocates claimed that the Internet could provide new forms of horizontal

and vertical communication, which had the capacity: (i) to broaden the range of

pluralistic voices heard in the public sphere and (ii) to facilitate new forms of

interactivity and deliberation, thereby (iii) widening the pool of political participants.

The majority of work focusing on the context of election campaigning has studied the impact

of the Internet from an institutional perspective. The early works found little evidence that the

Internet was having any impact beyond strengthening existing patterns of participation

(Bimber 1998; Corrado 2000; Davis 1999; Davis and Owen 1998; Hill and Hughes 1998;

Kamarck and Nye 1999). The evidence led Margolis and Resnick (2000) to articulate an

enduring perspective of the Internet as perpetuating what they describe as ‘politics as usual’.

In rebutting the cyber optimist perspective they observe that “Far from remaking American

politics, the development of cyberspace, and particularly of the WWW, seems more likely to

reinforce the status quo” (Margolis and Resnick 2000: 54).

There is thus a perspective of standardization in web-based campaigning that has revolved

largely around producing an online variant to the offline campaign that reinforces the key

messages of the party or candidate. Variations, where they occur, are explained by national

characteristics such as the technological infrastructure, the institutional arrangements, legal

provisions governing elections and the political culture; voting systems for example shape the

level of localism and individualization in electoral campaigns; party and presidential systems

govern the level of personalization. Overall however, the Internet is shaped by the political

logic of the campaign and not the reverse. Any moves towards more sophisticated and

participatory uses of the Internet have been incremental. In her review of European party uses

of the Internet, Norris (2003) demonstrates the web offering greater pluralism between parties

but largely the communication retains the top-down political logic of campaign

communication. As technologies have evolved, and interactivity facilitated in myriad ways,

parties have begun to offer more means for participation; but these would appeal only to the

most active, the already converted, rather than serving to attract citizens into engaging with

politics, any specific party, or have a deeper cognitive involvement in an election contest.

These findings have largely held true in the era of Web 2.0.

Web 2.0 as a term is fairly meaningless, however it acts as a useful heuristic when

demarcating the era of static communication, where content was largely created by those with

a specific skills-base, and that of the Internet as a vast architecture of participation. Of course

various forums, the innovative publishing platform Blogger and email permitted wide ranging

forms of publishing and interaction; yet social networking, video sharing, and indeed the

encouragement of over-sharing, has grown since 2005. Parties were initially slow to colonise

these new participatory architectures (Jackson and Lilleker 2009) but fears of being left

behind during election contests ensured that campaigns moved into the communities offered

by Facebook, StudiVZ, Hyves and the myriad other spaces used for social communication

(Lilleker and Jackson 2011). While the interactive opportunities offered by the 2008 Obama

campaign for the US presidency overshadowed moves towards more participatory styles of

campaigning (Harfoush 2008; Johnson 2009; Panagopoulos 2009), Web 2.0 tools are now de

rigueur and thus it is more common to find campaign websites balance informing with

interactive opportunities (Lilleker and Jackson 2011) than was previously the case (Kluver et

al. 2007). The increased use of interactive features means parties and candidates offer a more

uniquely online experience within their web-based presences, an experience that informs and

engages while offering opportunities to have a say and become involved (Lilleker et al.

2011). Yet, the strictures of political logic remain to govern online campaigning; interactivity

is a trade-off in order to use gimmicks and modern platforms, few parties or candidates see

value in entering into debates with visitors to their websites and largely avoid uncontrolled

interpersonal contact (Stromer-Galley 2000; Lilleker et al. 2010).

There is little research that provides a clear insight into the perspectives of the citizen on

matters of online campaigning. Norris (2003), in offering the “preaching to the converted”

thesis counters the argument that the Internet would usher in a greater political pluralism

across society. More recently, though focusing on the US context, Hindman (2009) found the

most read weblog authors replicate offline sources of influence. The tools may be there to

facilitate pluralism, and “it may be easy to speak in cyberspace”, but Hindman maintains that

“it remains difficult to be heard” (Hindman 2009: 142). Yet these patterns may well also be

undergoing an process of change. Research on online information seekers during the 2011

Polish election found four discrete patterns of participation: traditional offline activism,

creating content on weblogs and microblogs, monitoring the online political environment

through visiting a wide range of websites and, perhaps importantly in identifying new forms

of participation: the range of sharing and liking activities facilitated through social networks

for example (Koc-Michalska, Lilleker and Suroweic 2012). The findings chime with research

data from US elections which shows the digital divide defined by economic resources,

education and gender narrows for social networking site users. Pew Research Center data

demonstrates those on lower incomes are likely to make small donations to campaigns and

the lower educated are as equally likely to participate by commenting, sharing, and liking as

their better educated peers (Smith et al. 2009). This would seem to suggest there is some form

of Internet-based impact that is positive for widening political participation.

Engagement with election campaigns is also being reshaped due to the accessibility of

information via the Internet. Reviewing Internet use during the 2010 mid-term US elections,

Smith (2011) found 58 percent seeking political information online, and for 32 percent it was

the only information source. More importantly, 53 percent performed at least one action

classified as civic engagement. Evidence suggests the Internet is also increasing the level and

type of participation around election campaigns. Online users report they feel it easier to

connect with others who share their political views, suggesting that people seek out

communities of like-minded individuals. Yet this does not suggest the online election

environment consists of ideologically insular cyberghettoes, as suggested by Sunstein (2007).

Online users report the Internet facilitates exposure to a wider range of political views than

they can get in the traditional news media, perhaps reflecting on the highly partisan US media

environment, so broadening knowledge and facilitating a more active public sphere (Smith et

al. 2009). However, the parties and candidates need to provide tools that inform, engage,

mobilise and permit interaction with visitors; and arguably there are benefits to this.

Campaigns are clearly enhanced through working with a community of activists, it makes for

a more dynamic, involving and persuasive atmosphere at the very least. While Barr (2009)

steers clear of suggesting a causal link between Obama’s 2008 campaign strategy and his

victory, she notes the use of text messaging, Facebook, MySpace, email, and interactive

platforms facilitated the establishment of a range of youth-led pro-Obama organizations that

gave the campaign greater reach, increased voter registration among low-involvement groups,

and mobilised young and minority group voters on election day. Further evidence suggests

the more active a campaign is, the more engaged a following they gain; this is also the case

when measuring the impact of candidates’ or parliamentarians’ communication in terms of

gaining fans, followers, or getting content shared within networks (Jackson and Lilleker

2011; Lilleker and Koc-Michalska 2013). The more posts made to Facebook or Twitter, the

more weblog posts authored, the more likely they are to reach a wide audience and encourage

participation. It is further argued that measures of attention received can be used as a

predictor of votes. Tumasjan et al. (2011), analysing sentiment in public posts to the Twitter

platform, notes that such tools “can be a valid indicator of the political landscape off -line”

(Tumasjan et al. 2011: 414). Studies have also isolated web campaigning as a variable in

testing for explanations for candidate victories. When weighted for both political, party and

mainstream media coverage, there remains a significant effect from being online and, where

it is tested for, having the more interactive profile (Koc-Michalska, Lilleker & Suroweic

2012; Sudulich et al. 2012). These are still indicators from specific case studies but may offer

some evidence that as more people have unfettered Internet access, and become reliant upon

the Internet for information, that it may also become one driver for voter decision making

and, at the very least, to not have an online presence could be electoral suicide. Whether that

presence should also grab attention, engage and involve a visitor and offer opportunities to

participate is a matter for further investigation.

Methodology

This chapter seeks to measure evolution in online behaviour during election campaigns in

two ways. Firstly we focus on the behaviour of the French presidential candidates and their

respective parties in offering experiences to their online visitors. This research also explores

the way that different visitor groups were served by the provision of certain features and

content. The questions relating to the behaviour of candidates and parties are answered

following a content analysis procedure that has been the staple of Internet research since the

1990s. The coding scheme was developed by Gibson and Ward (2000) and has helped create

a rich picture of the role that the Internet has played within election campaigns globally

(Ward and Gibson 2003; Tkach-Kawasaki 2003; Foot and Schneider 2006 ; Conway and

Donard 2004; Schweitzer 2005; Strandberg 2006; Kluver et al. 2007; Ward et al. 2008). The

coding scheme has been expanded and updated over the years, in particular to accommodate

Web 2.0 tools and platforms (Lilleker and Jackson 2011), as well as formulae for categorising

features in order to understand the overarching strategies behind the design of online

campaign presences.

We analysed the websites and social networking profiles of all ten candidates and their

parties. The coding scheme contained a list of 92 discrete items and features which were

coded as present or absent. Where we identified comments, blog posts or other items we

produced a count of the number which appeared one month before the first round of the

French Presidential election (22 March to 22 April 2012) . Items and features are categorised

in terms of the experience they offer, the audiences they serve best and their contribution to

the strategy of the campaign. Our analysis tests for levels of use of features classified as

belonging to the Web 1.0 or Web 2.0 eras and then the extent they simply provide

information, provide an engaging or entertaining experience, are designed to mobilise

supporters or offer interactivity. We also assess whether the candidates and parties target

browsers with information and gimmicky but entertaining features, those interested in

political or policy issues, supporters or existing campaigners. Finally we focus on strategy,

assessing the extent that the candidates and parties focus on e-representation, focusing on

service to the nation and their past or future record, or a sales strategy that employs extensive

persuasive communication. In order to compare parties we develop average online

performance (AOP) scores for each batch of features. The AOP score was calculated by

initially counting the number of features present for each category to create an overall mean

per strategy. We then divided the mean score for each strategy by the maximum possible

score in that category. This is a common method of producing clear and transparent

comparisons used in a range of previous studies (see for example Farmer and Fender 2005;

Larsson 2011)

The second part of the chapter relates to the online survey run among sample representative

for French internet users (N=2630)1. The study was run before the first wave of presidential

elections in March 2012 focusing on online political activities, usage of different information

sources and electoral behaviour.

Online campaigning in France 2012: understanding the supply side

French citizens wishing to follow the online campaigns of the presidential candidates and

browse the candidate and party websites for information to aid their voter choices would have

found a dynamic environment offering a wide range of information in engaging formats as

well as opportunities to both get involved in the campaign and interact with other citizens and

the campaign team itself. Whether the opportunities promised more interactivity than was

available, for example whether emails, comments on weblogs or social media platforms were

ever answered by the campaign team, is a moot point. But all candidates and parties offered a

range of opportunities. In order to understand the supply side we focus on four elements of

the campaign, sophistication determined by the era of features, communication style,

audiences targeted and their strategy. All characteristics can be compared as they are counted

as average online performance (AOP) scores. Table 1 offers an overview of the campaign

environment to provide a sense of what the politically interested browser, perhaps seeking

ways to finalise their voting decision might find.

Table 1 about here please

Candidate websites, including any linked features, had most Web 2.0 era features and were

designed for the mobilisation of their supporters; the second largest number of features

offered some form of interaction. Information, including information in engaging formats

was clearly present but features that offer information no longer dominate the websites of

candidates or parties. Candidate sites targeted campaign activists, supporters and then

provided information for issue activists and professional information seekers, browsers were

least focused upon. Yet, despite not targeting browsers, the sites adhered slightly more to a

sales strategy than one conforming to e-representation. However the AOPs are not markedly

different so one can also argue that the website offered a balance of experiences for a range

of visitors. Party websites were built using more Web 1.0 era features and contained on

average less features. However the largest number of features permitted interaction, features

that offer straight information were however the second priority. Again the professional

information seekers and activists were well served, suggesting a much more policy heavy

focus; but once again the overall experience would have been more persuasive than

representational. The AOP scores are broken down further to compare the online

campaigning styles and strategies for each party and candidate; the results are displayed in

Table 2.

Table 2 about here please

1 Mediapolis, ANR project 2008 – 2012, dir. Thierry Vedel, CEVIPOF, Sciences-Po Paris

Overall the data shows that all sites offered a mixture of features and so catered for as wide

an audience as possible. While informing, and providing information in engaging formats

such as videos remain key features, both candidate and party websites are far more likely to

focus on mobilising supporters who visit their websites; this may reflect recognition that

supporters, whom Norris (2003) refers to as the converted, are the most likely visitors to

official websites. Importantly, and contrasting with the data from studies conducted prior to

social media taking off (Kluver et al. 2007), offering interactive opportunities is no longer

treated with caution it would seem. The interactive communication strategy is aided by the

adoption of social networking, with nine out of ten candidates and half the parties having a

Facebook profile and all candidates and eight out of the ten parties having an active Twitter

account. Parties and candidates also allow sharing of content via social networks, this by

default permits adaptation and comment on their messages if the user desires adding a more

public interactive dimension that is completely beyond the control of the host. The provision

of such opportunities is high. Table 3 shows the averages for content created by party and

candidate sites and profiles, a comparison of the major, minor and fringe parties as well as the

individual scores for the top two candidates.

Table 3 about here please

Overall we find candidates and parties, but particularly candidates, creating a large amount of

content that could be interacted with, shared and liked. Many candidates and parties created

what we have referred to elsewhere as hybrid sites (Lilleker and Koc-Michalska 2013),

websites that incorporate weblog features mirroring the site of Obama at the 2008 US

election contest. Interestingly, neither Hollande nor Sarkozy chose this route. Instead they

seemed to focus more on social networks with Hollande creating most content for Facebook

and Sarkozy having a highly proactive Twitter strategy. Disparity between the candidates’

levels of visible online support can be partially explained by the fact that Sarkozy had a

strong historical presence on Facebook (built during his presidency) however his uptake of

Twitter was a far more recent innovation created a day before announcing his candidacy

(February 2012) and was not updated after the second round of voting (May 2012). The

disparity in Facebook fans and Twitter followers with the amount of posts is interesting.

Contrasting the view that support online mirrors support offline, Sarkozy’s 600,000 strong

following on Facebook dwarfs that of Hollande, with their Twitter following being close but

seeing Hollande emerge ahead; but contrasting with other studies (Tumasan et al. 2011) we

suggest neither can be treated as verification of a mass support beyond these environments

however.

As Table 3 also shows the overall sophistication, in terms of total number of features, and the

levels of proactivity in creating content for social media, fluctuates comparatively to the

levels of support the candidate and party have had historically (based on vote share in 2007);

a correlation of .647 significant at the 0.01 level. Previous high or low levels of support may

be indicative of resources available and thus reinforce the argument that offline inequalities

are mirrored online (Margolis and Resnick 2000).

In summary, strategically there is something of a gulf between the candidate and party sites in

terms of the latter being on average more information heavy and designed for browsers. Yet,

for the sites of the top two candidates, Sarkozy and the UMP website and the websites of

Hollande and the Parti Socialiste, there is more similarity in the level of overall sophistication

than difference. The competition between left wing candidates seemed to have been driving a

focus on mobilising supporters, reflected by the high averages for feature use in the category

of mobilisation and targeting of supporters by both Holland and Bayrou. The right wing party

candidates were, however, not far behind in mobilising supporters or offering visitors

opportunities to interact. Focusing on the candidates who went into the second round, and

indeed comparing the online strategy of winner Hollande to his opponent Sarkozy, we find

Sarkozy offering a highly sales-oriented campaign aimed at browsers and issue activists.

Hollande, as indicated earlier focused mostly on mobilising his supporters but also balanced

the sales dimension with an e-representation strategy. The Hollande website also contained

more features, including information, suggesting his campaign used the website as a space

where visitors could learn about him and what he might offer as president as well as building

a hub for his activists. Sarkozy promoted himself as the president that could build a strong

France, a more national focused but de-personalised message that seemed aimed at building a

coalition across the French right. Unlike Hollande, the Sarkozy campaign was cautious in

providing an inclusive space for visitors to find a range of messages or offer their comments.

Online information seeking and voter decision making: understanding demand

The online environment was certainly vibrant during the 2012 French presidential election

contest, not only due to the promotional activities of the candidates and the parties but also

based on the self-reported activities of French citizens. The Mediapolis project involved

asking citizens whether they performed any of sixteen forms of participation; eight were

offline, six related to accessing information online and the final five to activities that could be

regarded as social using Web 2.0 tools or platforms such as Facebook. The full breakdown is

shown in Table 4. Headline findings suggest mixed levels of participation; but regression

analysis indicates that there is a relationship between being highly active offline and being

highly active online. The majority who are active online are actually passive, however.

Searching for and accessing information dominates, but not always from official sources.

Political satire is the most popular media; this could also include some of the material under

the category of political videos, though the latter may also include official candidate

communication. Candidate websites and political weblogs do not fare badly, but the more

active discussions in forums are the preserve of a minority. The use of social networks by

candidates may, however, pay dividends as it seems a considerable number are willing to

share political information via these platforms; although one cannot tell what that content is

and much of this may be unofficial and of a satirical nature. Some receive political

information, probably shared by other users, but few post content; even less use Twitter to

either receive or send content.

Table 4 about here please

Some studies have shown that the Internet now rivals the press and radio as a key source of

political information and this is also true for France, but television remains the dominant

information source. In terms of citizens who sought support for their voter choices, measured

on the basis of having switched their vote between rounds one and two, radio listeners were

the most likely (31%), Internet users second (29%) and television viewers third (26%). The

least likely to change their vote decision are newspaper readers (17%). But this does not

indicate that the sites we are focusing on are of importance. Only 15% of those who switched

their vote visited the website of a party or candidate; 16% engaged with information on social

media. Therefore, demand for these sites is actually quite low. 15% engaged in some form of

activity on social media; these may be a significant number but they are also a minority2.

Digital divides, however, may be decreasing. Hindman (2009) argued that those political

active online are likely to be white, male, highly educated and from a professional

background. While we have no data on race, using regression analysis we discovered some

interesting and sometimes unexpected patterns. Female citizens were, as expected, less likely

to visit party or candidate websites, post to weblogs or participate in political forums. But, in

the case of social networks, there is no gender imbalance. Similarly it is the least educated,

those attaining less than a Baccalaureate, who are politically active on Facebook. This group

is also most likely to use forums on the contrary there is no education effect for visiting

websites, reading weblogs or using Twitter. Furthermore there is less inequality across the

age range. Age does not predict likelihood of visiting websites or Twitter usage; although the

young are still more likely to read political weblogs and be politically active on Facebook.

These data suggest that the online environment should no longer be treated as an exclusive

medium in terms of demographics.

The most important point to note is that while 56.6% of those online in France went online to

seek information on the election, few visited the websites of political candidates or parties.

The numbers visiting the websites or social media platforms of candidates or parties were not

even a majority of those who claim to have high interest in politics. Of that group 57% visit

official websites and 64% are politically active on Twitter and 55% on Facebook. Those who

are most likely to visit candidate and party websites are a fairly specific group which clearly

adhere to the notion of them being already converted. While we cannot determine causality,

those who visit official websites are most likely to donate money to a political campaign, and

distribute posters or flyers as part of being an activist. These respondents have a high level of

political interest and are most likely to discuss politics with others; so not only being loyal

supports but advocates. Reflecting Hindman’s (2009) concerns about an online digital divide,

they are also most likely to be male, older and have higher levels of education, and believe

politicians care about people as opposed to viewing politics and politicians cynically. In

contrast, we can suggest that social media reaches a small number of people that websites do

not. Of those with little or no interest in politics, 17% ever visited the website of a party or

candidate but 19% took part in some form of political activity on Twitter and 25% via

Facebook. While still a small number it may suggest social media may act as a gateway into

2 For further details on the study please see Koc-Michalska and Vedel 2013.

political participation, even if still for a minority. Our survey data indicates a general

importance of the Internet but not necessarily of party or candidate websites. However, some

Internet users, those in younger age groups, females and those with lower education levels

may well be drawn into becoming interested not by the activities of parties or candidates per

se, rather by the sharing and liking activities of their online ‘friends’ who have a higher

interest in politics. It is suggested that through harnessing the labour of activists, encouraging

them to undertake small tasks on social media sites such as sharing a link or liking a message,

the party or candidate can extend their reach (Lilleker 2013). While it is impossible to

provide information about the senders of information, we know that 34% (of those having

social network profile, 62% of respondents) received political information via social

networks and 23% posted political content. Among those having Twitter accounts (11% of

the internet users) 47% received political information and 32% sent tweets about politics.

Even if this group of active users is not particularly numerous they may play an important

role in shaping the opinions of others.

While an understanding of the main media of choice, as well as sources for gaining political

information, offer some insights into where candidates and citizens may meet; insights into

what citizens who go online for political information seek are more important. In responding

to the question 'what information do want about the candidates’ the majority (73%) respond

that they search for the political programme of candidates. Only a minority search for

information about the private lives of candidates (7%), while 18% are interested in

information about campaigning techniques. However, breaking this down by platform, there

are interesting differences which are shown in Table 5. The visitors to candidate or party

websites are most interested in policy (70%), followed by readers of political weblogs

(69.7%) and, surprisingly, Facebook users who receive political information and create (64%)

or share content (62%). Twitter users are least likely to be seeking information on the

programme and policies of a candidate, although they still represent over 50% of users of the

platform, rather Twitter users seem to favour more personalised messages than users of any

other platform. The averages for interest in information on campaign tactics does not differ

vastly across platforms, although those who discuss politics on Forums (31%), followed by

Twitter users seem most interested in this type of information.

Table 5 about here please

While it is impossible to isolate activists within the Mediapolis survey, identifying the vote

switchers and the actions they have taken that may have aided them to choose a candidate to

support offer further indications of what those who go online seeking political information

feel it is important to find. Again, among Internet users these were a minority. Also there is

no indication that Internet users were more likely to be persuaded due to having access to a

wide range of information offering multiple perspectives. In fact, our data shows that Internet

users were even less open to receiving new information and preferred online content that

reinforced their existing political beliefs. Yet, we can still view this small group as important,

representing as they do 27% of the respondents, all of whom were forced to reconsider their

choice between the first wave of our survey (run in March) and the vote declaration (asked in

the last wave run after the election) The full breakdown of factors that aided voter choices,

with percentages, are shown in Table 6. The interesting points to note are that the most

important factor, which was argued to be key to aiding voter choice by 50% of respondents

is purely related to access to information regarding ‘Candidates ideas or policy proposals’.

While perhaps more subjective judgments, ‘Their ability to have an impact in French politics

and society’ (49%), ‘Their ability to deal with the economic crisis’ (34%) and ‘The candidate

was the least worst option’ (9%) equally necessitate accessing detailed information to aid

comparisons.

Table 6 about here please

In summary, respondents to the Mediapolis survey demonstrate that the online environment is

an important space during elections. The Internet is one space to turn for information, in

particular for some who are seeking information to help them make a decision on whom to

support. Candidate and party websites and their profiles on social networks all have an

audience, albeit a limited one both in terms of absolute numbers and demographics. The

official websites tend to attract the converted; social media, in contrast, provide spaces for

learning and sharing to those who are not highly interested in politics and do not usually

engage political news, official sites or weblogs. The role of social media sites, in particular

Facebook, as a space for widening political participation, conceptually and practically, in

France mirrors wider tendencies among young, non-politically active, citizens (Koc-

Michalska, Lilleker and Suroweic 2012). However, the information sought may be driving

online users to non-official sites as opposed to drawing them to the platforms of candidates

and parties. There may be elements of post-hoc rationalisation, as reflected by some

inconsistencies in findings such as the demand for information and the most popular features

containing humour or being of a satirical nature. That caveat aside, however, largely what

was sought was information about the proposed policies and programmes of the candidate;

there was little demand for personalised content and none for the ability to ask questions or

participate beyond active information seeking.

Supply versus Demand: understanding campaign environments

As is the case with candidates and parties in most advanced and industrial democracies, the

2012 French presidential candidates embraced the Internet as a communication tool.

Alongside the traditional provision of information there were features designed to engage and

entertain visitors to their websites, mobilise them into forms of activism and features that

permitted interaction with other visitors and the campaign. The use of social media added a

further dimension to both mobilisation and interactivity, providing as it does simple ways to

participate and act as an advocate as well as sharing thoughts related to posts from the

campaign. Social media appear to provide a gateway into some forms of political

participation, some of which are simplistic, but that may actually have impact upon the

participant, who might like a picture or statement from a candidate, and their online contacts

who view the item that has been liked. Thus elections now link well into the online

communication ecosystem and there is a broad conversation, taking place across numerous

websites and platforms, about the election.

Many citizens use this ecosystem, though perhaps not in the way that campaign strategists

desire. There is much greater interest in satire than the website of the candidate. There is also

much more general information searching that directly going to the sites of candidates or

parties. More importantly, however, is the nature of the information sought. These active

information seekers claim to want information on the party platforms and policies; a minority

seek information about the personality of the candidate or the conduct of the campaign. One

can enquire, therefore, how well served these information seekers are. Candidate and party

websites tend to follow the norms of election campaigning, presenting a simplified and

persuasive perspective of their own platform. The former may be of use, as many are likely to

seek simple information as opposed to lengthy manifestoes. However, it is argued that

citizens do not like overly persuasive communication and are likely to become cynical in

reaction to political sales pitches and reject both the message and the sender.

We might suggest, based on the above data, that Partie Socialiste candidate and eventual

winner Francoise Hollande had the best online communication strategy. While highly Web

2.0 he presented slightly more information and delivered it alongside a much more engaging

and entertaining experience. He also focused more on an e-representation message, though

still outweighed by sales, which may have given a greater sense to his visitors of what he

offered as a candidate. Of course we would not claim this was in anyway pivotal in

Hollande’s campaign, more perhaps reflective of a broader campaign strategy within which

the online communication is integrated and embedded. However, we tentatively claim that

Hollande provided a better online environment for the French ‘Internaut’, particularly vote

switchers, and this may have won him some support among that group.

The much broader question our data raises is whether online users really want to interact with

the campaign? Jackson and Lilleker (2009) have commented that interaction on official

political websites, weblogs or social media profiles tend to more resemble graffiti that

interjections into a conversation. Hollande received comments to one post saying “well said”

and “you must be crazy”, Sarkozy, publicising a speech he had delivered “you were fantastic”

and “you are an arsehole”; neither encourage discussion. Such comments call them graffiti or

flaming become part of the campaign communication ecosystem; there for anyone to access

though likely to go unread by anyone when there are significant numbers of comments. But

there are few examples of a direct question to the candidate made via these public profiles.

Therefore, despite many arguments regarding the enrichment of democracy through enhanced

interactivity, there is little evidence of democratic engagement taking place. There is equally

less evidence this is desired. Information, presented in engaging ways, perhaps with a greater

focus on e-representation than sales would appear the ideal product, scattered across a range

of platforms to be accessible to the fragmented online media audience. Whether a more

accessible, interactive and responsive candidate or campaign is desired is a moot point, but it

appears that demand is at least fairly low.

References

Barr K (2009) A Perfect Storm: The 2008 Youth Vote. In Johnson D (ed.) Campaigning for

President 2008: Strategy and Tactics, New Voices and New Techniques, London, Routledge,

p 105–25.

Baym N (2010) Personal Connections in the Digital Age, Cambridge, Polity.

Bimber B (1998) The Internet and Political Transformation: Populism, Community and

Accelerated Pluralism. Polity XXXI (1), p 133–60.

Conway M, Donard DG (2004) An Evaluation of New Zealand Political Party Websites.

Information Research, 9(4): Online. Available HTTP: < http://informationr.net/ir/9-

4/paper196.html >. Accessed September 2, 2012.

Corrado A (2000) Campaigns in Cyberspace: Toward a New Regulatory Environment.

Washington, DC, The Aspen Institute.

Davis R, Owen D (1998) New Media and American Politics. New York, Oxford University

Press.

Davis R (1999) The Web of Politics. Oxford, Oxford University Press.

Dertouzos M (1997) What Will Be: How the New Information Marketplace will Change our

Lives. San Francisco, CA, Harper.

Farmer R, Fender R (2005) E-parties: democratic and republican state parties in 2000. Party

Politics 11, p 47-58.

Foot KA, Schneider SM (2006) Web Campaigning. Cambridge, MA, MIT Press.

Gibson R, Ward S (2000) British Party Activity in Cyberspace. In Gibson R, Ward S, (eds)

Reinvigorating Government? British Politics and the Internet, Aldershot, Ashgate, p 155–

207.

Harfoush R (2008) Yes We Did: An Inside Look at How Social Media Built the Obama

Brand. Berkeley, CA: New Riders.

Hill KA, Hughes JE (1998) Cyberpolitics: Citizen Activism in the Age of the Internet.

Lanham, MD, Rowan & Littlefield.

Hindman M (2009) The Myth of Digital Democracy. New York, Lexington.

Jackson N, Lilleker D (2009) Building an Architecture of Participation?: political parties and

Web 2.0 in Britain. Journal of Information Technology and Politics, 6 (3/4), p 232-250.

Jackson N, Lilleker DG (2011) Microblogging, Constituency Service and Impression

Management: UK MPs and the Use of Twitter. Journal of Legislative Studies 17(2), p 86–

105.

Johnson D W (2009) Campaigning for President 2008: Strategy and Tactics, New Voices and

New Techniques. London, Routledge.

Kamarck EC, Nye JS Jr (1999) democracy.com? Governance in a Networked World. Hollis,

NH, Hollis Publishing.

Kluver R, Jankowski NW, Foot KA, Schneider SM (2007) The Internet and National

Election: A Comparative Study of Web Campaigning. London, Routledge.

Koc-Michalska K, Lilleker DG, Suroweic P (2012) Talking to the politically engaged:

exploring the Polish electoral blogosphere. Paper presented at the IPSA 2012 World

Congress of Political Science, Madrid, July 8th

.

Koc-Michalska K, Lilleker DG, Baranowski P, Suroweic P (2012) Poland’s 2011 online

election campaign: new tools, new professionalism. Paper presented at Chasing the Digital

Wave: International Lessons for the UK 2015 Election Campaign, London, November 22nd

.

Larsson AO (2011) Extended infomercials” or “politics 2.0″? A study of Swedish political

party Web sites before, during and after the 2010 election. First Monday, 16: 4, 4 April.

Available online at

http://www.uic.edu/htbin/cgiwrap/bin/ojs/index.php/fm/article/view/3456/2858

Lilleker DG (2013) Empowering the Citizen? Political communication, coproduction and the

harnessed crowd. In, Scullion R, Gerodimos R, Jackson D, Lilleker DG (eds) The Media,

Political Participation and Empowerment. London, Routledge

Lilleker DG, Jackson N (2011) Political Campaigning, Elections and the Internet: Comparing

the US, UK, Germany and France. London, Routledge.

Lilleker DG, Koc-Michalska K (2013) Online political communication strategies: MEPs, e-

representation and self-representation. Journal of Information Technology and Politics,

forthcoming.

Lilleker DG, Pack M, Jackson N (2010) Political Parties and Web 2.0: The Liberal Democrat

Perspective. Politics, 30(2), p 105–12.

Margolis M, Resnick D (2000) Politics as Usual: The Cyberspace Revolution. Thousand

Oaks, CA, Sage.

Negroponte N (1995) Being Digital. New York: Knopf.

Norris P (2003) Preaching to the Converted? Pluralism, Participation and Party Websites.

Party Politics, 9 (1), p 21–45.

Panagopoulos C (2009) Politicking Online. Lundon: Rutgers.

Press AL, Williams BA (2010) The New Media Environment. Oxford, Wiley-Blackwell.

Rheingold H (1993) The Virtual Community: Homesteading on the Electronic Frontier.

Reading, MA, Addison-Wesley.

Schweitzer EJ (2005) Election Campaigning Online: German Party Websites in the 2002

National Elections. European Journal of Communication, 20(3), p 327–51.

Smith A (2011). Why Americans Use Social Media. Available at

<http://www.pewinternet.org/Reports/2011/Why-Americans-Use-Social-Media.aspx. >

Smith A, Schlozman KL, Verba S, Brady H (2009) The Internet and Civic Engagement.

Available at < http://www.pewinternet.org/Reports/2009/15--Th e-Internet-and-Civic-

Engagement. aspx.>

Strandberg K (2006) Parties, Candidates and Citizens On-Line Studies of Politics on the

Internet, Abo Academy University Press: Abo, Finland.

Stromer-Galley J (2000) Online Interaction and Why Candidates Avoid It. Journal of

Communication, 50(4): 111–32.

Sudulich L, Wall M (2012) The Effects of Internet Use on Voters' Electoral Uncertainty.

Presented at ECPR Joint Sessions of Workshops, Parties and Campaigning in the Digital Era

University of Antwerp, Belgium, April 22nd.

Sunstein C (2007) Republic.com. Princeton, NJ, Princeton University Press.

Tkach-Kawasaki L (2003) Politics@Japan: Party Competition on the Internet in Japan. Party

Politics, 9(1), p 105–23.

Tumasjan A, Sprenger TO, Sandner PG, Welpe IM (2011) Election Forecasts with Twitter:

How 140 Characters Reflect the Political Landscape. Social Science Computer Review,

29(4), p 402–18.

Ward S, Gibson R (2003) Online and On Message? Candidates Websites in the 2001 General

Election, British Journal of Politics and International Relations, 5(2), p 188–205.

Ward S, Owen D, Davis R, Taras D (eds) (2008) Making a Difference: A Comparative View

of the Role of the Internet in Election Politics. Lanham, MD, Lexington Books.

Table 1: Understanding the Campaign Environment Online

Feature Category Candidate Website Party Websites

Web 1.0 .534 .544.

Web 2.0 .586 .491

Informing .483 .535

Engaging .448 .477

Mobilising .667 .520

Interacting .663 .589

Targeting Browsers .475 .495

Targeting Issue Activists .563 .574

Targeting Supporters .520 .517

Targeting Campaign Activists .566 .534

E-Representation Strategy .511 .458

Sales Strategy .607 .582

Overall Performance .554 .530

Table 2: Comparing the Campaign Styles and Strategies of the Candidates and Parties in France 2012

Web 1.0 Web 2.0 Informing Engaging Mobilise Interact Browser Issue

Activist Support Campaign E-Rep Sales

Candidate (C)

Party (P) C P C P C P C P C P C P C P C P C P C P C P C P

Sarkozy/UMP .62

9

.88

6

.54

5

.59

1

.37

5

.72

5

.35

5

.67

7

.80

0

.80

0

.68

4

.73

7

.60

0

.80

0

.63

2

.78

9

.53

7

.70

7

.53

2

.76

6

.44

7

.73

7

.67

9

.82

1

Hollande/PS .57

1

.68

6

.72

7

.59

1

.42

5

.62

5

.61

3

.64

5

.93

3

.53

3

.73

7

.78

9

.55

0

.75

0

.63

2

.73

7

.56

1

.63

4

.70

2

.61

7

.65

8

.57

9

.75

0

.78

6

Bayrou/MoDe

m

.71

4

.45

7

.72

7

.72

7

.45

0

.47

5

.45

2

.48

4

.93

3

.60

0

.84

2

.78

9

.55

0

.60

0

.63

2

.63

2

.63

4

.58

5

.70

2

.55

3

.60

5

.47

4

.67

9

.60

7

Le Pen/FN .68

6

.77

1

.63

6

.68

2

.55

0

.67

5

.58

1

.61

3

.73

3

.80

0

.78

9

.73

7

.65

0

.65

0

.73

7

.63

2

.61

0

.68

3

.66

0

.74

5

.60

5

.57

9

.78

6

.64

3

Melanchon/Fd

G

.40

0

.51

4

.59

1

.45

5

.65

0

.40

0

.54

8

.45

2

.53

3

.60

0

.68

4

.47

4

.45

0

.35

0

.63

2

.36

8

.53

7

.39

0

.53

2

.51

1

.52

6

.34

2

.60

7

.46

4

Joly/EE .65

7

.48

6

.54

5

.59

1

.42

5

.52

5

.38

7

.58

1

.86

7

.60

0

.63

2

.73

7

.40

0

.55

0

.47

4

.63

2

.48

8

.58

5

.61

7

.53

2

.55

3

.55

3

.60

7

.71

4

Dupont-

Aignan/DLR

.45

7

.28

6

.81

8

.27

3

.52

5

.27

5

.58

1

.19

4

.66

7

.26

7

.84

2

.31

6

.65

0

.20

0

.68

4

.31

6

.63

4

.34

1

.66

0

.31

9

.60

5

.18

4

.67

9

.28

6

Poutou /NPA .45

7

.42

9

.40

9

.45

5

.35

0

.60

0

.29

0

.51

6

.46

7

.26

7

.52

6

.52

6

.35

0

.45

0

.26

3

.73

7

.39

0

.43

9

.40

4

.46

8

.34

2

.44

7

.42

9

.60

7

Arthaud /LO .31

6

.51

4

.36

4

.27

3

.47

5

.60

0

.22

6

.32

3

.20

0

.33

3

.31

6

.36

8

.20

0

.30

0

.42

1

.52

6

.34

1

.46

3

.31

9

.40

4

.28

9

.39

5

.28

6

.50

0

Cheminade

/SeP

.45

7

.51

4

.50

0

.27

3

.60

0

.45

0

.45

2

.29

0

.53

3

.40

0

.57

9

.42

1

.35

0

.30

0

.52

6

.36

8

.46

3

.34

1

.53

2

.42

6

.47

4

.28

9

.57

1

.39

3

Table 3: Content creation and responses from the audience

Website Weblog Facebook Twitter

Entries Comments Posts Comments Posts Comments Friends Tweets Followers

Overall Average 150.15 730.40 5.25 72.70 59.2 15325.95 55137.65 765.5 48418.8

Candidates 205.60 1243.10 10.50 145.40 58.7 28851.7 92961.5 616.3 75278.8

Hollande 999 0 93 134 66 74094 95520 806 245553

Sarkozy 325 0 0 0 27 66521 607863 1517 177262

Parties 94.70 217.70 0.00 0.00 59.7 1800.2 17313.8 914.7 21558.8

Parties and candidates

Major 396.75 0.00 23.25 33.50 112.5 38081.25 194177 1124.75 120328

Minor 86.70 1449.50 0.00 132.00 63.8 15343.5 31007.9 1002.1 46914.4

Fringe 91.50 18.83 1.00 0.00 16 126.5 2661 131.67 2986.67

Source: Website, weblog, Twitter: coders. Facebook: Sotrender.com

Note: for posts, comments and twits: data for period 22.03 to 22.04; for friends and followers data for 22.04.

Table 4: Percentages of citizens who reported engaging in some form of participatory activity

Percentage

Offline

Attend political meetings 17.9

Contact a politician 15.1

Sign a petition 53.8

Strike 31.9

Participate in a demonstration 36.0

Boycott a product or service 12.1

Distribute Flyers 9.2

Donate money to a political cause 6.9

Accessing Information Online

Search online for political information 56.6

Visit a candidate website 38.0

Watch a political video 47.5

Read a political weblog 34.9

Discuss politics in a forum 18.7

Read/Watch political satire 54.8

Use Web 2.0 tools and platforms

Send/Share political information 36.3

*Receive political information on social networks 34.4

*Post political information on social networks 22.7

*Receive political information on Twitter 46.6

*Post political information on Twitter 31.7

* Percentages of those with accounts only, for Facebook n = 1589; for Twitter n = 299

Table 5: Types of information sought by users of different platforms (%)

Information about the

programme or policy of a

candidate

Inside information on the

campaign and campaign tactics

Personal or private

information about the

candidate

Website Visitors 70.3 22.3 7.4

Weblog Readers 69.7 22.9 7.4

Discuss Politics of Forums 59.8 30.8 9.3

Receive information on Facebook 63.9 25.3 10.9

Post political information on Facebook 62.0 26.9 11.1

Receive information on Twitter 55.4 25.9 18.7

Post political information on Twitter 51.1 29.8 19.1

Table 6: Factors influencing voter decision making, as 1st and 2nd reason (%)

General

populationa

Vote changersb

Candidates ideas or policy proposals 26.6 49.7

Their ability to have an impact in French politics and society 23.9 49.3

Their ability to deal with the economic crisis 18.2 26.2

A willingness to listen to citizen’s concerns 9.1 24.2

The candidate’s personality 6.7 15.6

The candidate was the least worst option 6.2 9.4

The candidate had a good chance of winning the contest 3.7 4.9

I am closer to the party of the candidate 2.5 3.8 a percentage among all respondents that answer the question on vote decision N= 1696, b Percentage among the candidates who changed

their vote decision between March and the voting day in April N=274

Appendix

Nicolas Sarkozy UMP Union for a Popular Movement

Francois Hollande PS Socialist Party

Francois Bayrou MoDem Democratic Movement

Marine Le Pen FN National Front

Jean-Luc Mélenchon FdG Left Front

Eva Joly EE Europe Ecology

Nicolas Dupont-Aignan DLR Arise the Republic

Philippe Poutou NPA New Anticapitalist Party

Nathalie Arthaud LO Workers' Struggle

Jacques Cheminade SeP Solidarity and Progress