Article

French Politics (2008) 6, 1–22. doi:10.1057/palgrave.fp.8200139

Surfing to the Élysée: The Internet in the 2007 French Elections

Cristian Vaccaria

aDipartimento di Organizzazione e Sistema Politico, Università di Bologna, Strada Maggiore, 45, Bologna 40125, Italy. E-mail: cristian.vaccari@unibo.it

Top

Abstract

The internet was one of the most discussed communication innovations in the French 2007 Presidential and legislative elections. In order to assess the state of online campaigning in France, this article analyzes the characteristics of the websites of 12 presidential candidates and 10 national parties during the campaign. The data reveal that, despite the media hype, online electioneering in France is still at an intermediary stage, especially in terms of participation tools. Significant differences were found among candidates and, especially, parties. The gap between large and small parties is found to be greater than in most of similar country studies, thus providing new evidence against the internet's ability to level the political playing field. Distinctive patterns of online electioneering emerge between conservative and progressive parties and candidates.

Keywords:

elections, France, internet, parties, candidates, online campaigning

Top

Introduction

More than a decade since political parties and candidates began integrating the internet in their communication infrastructures, research on the role of ICTs in politics and campaigning has reassessed most of the early hypotheses, which claimed that the new media would revolutionize politics and democracy. The internet has not mobilized massive numbers of new political activists, but it has reinforced existing patterns of civic engagement and has made it easier and less costly for receptive organizations and social movements to mobilize their supporters. The ICTs have not radically altered power relationships, equalizing the playing field between large and small, resource-rich and resource-poor parties and organizations, but have widened the pool of competitors by reducing the costs of communication and organization and helping new political movements and entrepreneurs harness the energies of supporters unattached to traditional structures. Thus, rather than stimulating a transformation from representative to direct democracy, the new media have been gradually integrated in existing political practices by both parties and citizens (Ward and Vedel, 2006).

Conversely, most of the available research on Western parties' and candidates' adoption of ICTs has failed to detect a significant and widespread commitment to the participatory and mobilizing potential of the internet (Nixon et al., 2003; Ward et al., 2003). Political actors 'have simply transferred the existing framework from the real world to the virtual realm, using ICTs principally to improve the speed of organizational communication' (Bentivegna, 2006, 336). The most recurrent explanation for such a half-hearted approach is that parties and candidates are wary that their embracing of the democratic ethos of the new media could result in a loss of control over their message (Stromer-Galley, 2000). The nature of the internet as both an information channel and a participation device threatens to confuse the party's public self-presentation — which campaign strategists require to be as unified and consistent as possible — with internal debate and discussion — which, if genuinely open and egalitarian, can easily reveal differences and conflicts within the party. Moreover, Ward and Vedel talk about a 'fear factor within mainstream institutions' (2006, 220), as their failure to properly deal with the internet can create a backlash in the public sphere because of high media and citizen expectations and scrutiny.

Thus, as most parties engage in a trial-and-error process of adopting the new media, their approach can be summarized as 'integrating their offline message into their online campaign' (Ward et al., 2003, 19). When parties open channels for citizen engagement, they restrain users' options so as to minimize the possibility of unexpected events, spontaneous bottom-up initiatives, and muddling of the key campaign messages decided by the party elites and their consultants. As some scholars noted, 'The feedback sought almost always relates to positions that the parties have themselves pre-defined' (Cunha et al., 2003, 88). Moreover, interactivity requires elite response to users' contributions and online participation demands organizational support. Thus, many parties refrain from engaging their followers online due to lack of resources or unwillingness to deploy them to non-traditional media (Villalba, 2003, 127, 135).

These somber assessments are in line with Margolis and Resnick's (2000) famous assertion that 'politics as usual' would prevail online and that a process of 'normalization' would empty the internet of most of its innovative potential. A case in point is the fact that, while theoretically the new media could level the playing field between smaller and larger parties, in reality the latter usually have stronger ICT infrastructures due to their superior resources. Whereas this routinely proves true in a two-party system such as that of the United States (Margolis et al., 2003), where Republican and Democratic parties and candidates regularly outperform their minor counterparts, studies of various European multi-party systems (Cunha et al., 2003; Schweitzer, 2005) show that party size does not fully account for website quality and sophistication. For instance, a quantitative study on French parties' websites conducted in 2000 (Sauger, 2002) found that relatively small forces, such as the Parti Communiste Français (PCF), the Greens and the Union pour la Démocratie Française (UDF) outperformed the three largest parties: the Gaullist Rassemblement pour la Republique (RPR), the Parti Socialiste (PS), and the Front National (FN).

More generally, adoption of ICTs by political parties has been found to depend on three broad factors: the technological development, the socio-political environment (including electoral laws, types of elections, and party system structure), and internal variables (such as party resources, incentives, and philosophical orientation) (Nixon et al., 2003, 241). Thus, the factors that influence political actors' adoption of ICTs seem to be more nuanced than the mere availability of resources.

Another crucial force that influences the internet strategies of political actors is the morphology of the public of online politics. Research on both Europe (Norris, 2003) and the United States (Bimber and Davis, 2003) has found that users of political websites are by and large strongly interested in politics, more partisan than the rest of the population, and well integrated in other channels of political communication and participation. As a result, it is not reasonable to expect the website of a party or candidate to reach uncommitted voters hungry for issue enlightenment and deliberation. Rather, the core supporters who log on political websites are motivated by a desire of opinion reinforcement and active participation to the campaign. Therefore, in the multi-channel logic of contemporary electioneering, the main value of parties' and candidates' websites lies in their effectiveness in mobilizing supporters and providing avenues for engagement in old and new campaign activities, not in their limited ability to target and persuade undecided voters. Both political actors and citizens are thus reshaping the internet from a simple vehicle for information exchange, as it used to be considered in its early ages, to a tool for mobilization, organization, and participation. Such activities are no longer bounded within cyberspace, but increasingly extend to offline endeavors, as was the case with the so-called 'netroots' that came to the fore in the 2004 US elections (Williams and Tedesco, 2006).

Thus, while still limited to smaller and more politically idiosyncratic audiences than traditional mass media, the internet can nonetheless be seen as a relevant addition to the toolkits of parties and candidates in that it can help strengthen their organization and mobilization structures and reinvigorate their relationships with members and supporters, which a vast literature has found to be weakening over the last decades, with worrisome consequences for the health and quality of democracy (Dalton and Wattenberg, 2000). The evidence previously summarized, however, suggests that political actors have overall been reluctant in pursuing such a route. This research analyzes the 2007 French Presidential and legislative elections to understand whether parties and candidates confirmed or departed from the aforementioned tendencies.

Top

The French Public and Online Politics in 2007

The state of technological development of ICTs in France can be described as intermediate, both in terms of the percentage of the population who is connected to the internet and by comparison with other European countries. In 2006, 41% of French households had internet access, compared to 49% in the 27 countries of the European Union. Thirty percent of French households had broadband connections, on par with the EU average. Thirty-nine percent of the French population claimed to go online at least once a week, while 45% of EU citizens did.1 Industry estimates indicate that in April 2007 the number of French internet users neared 30 million, or 55% of people aged over 11 (Mediametrie, 2007).

Although the internet cannot compete with television as a campaigning outlet, it has achieved a solid position in the second tier of political media. According to a survey conducted after the Presidential election, 8% of the French public considered the internet as their main source of information and 21% mentioned it as either their first or second choice (CEVIPOF, 2007b, 27). Interestingly, the web became more relevant for voters as the campaign progressed, as only 5% had named it as their main source and 14% as one of the two main sources in Fall 2006 (CEVIPOF, 2007a, 29–30). Surveys taken during the election identified two main disparities in the composition of the French e-citizenry. The first is the digital divide in internet access between higher and lower social strata and between younger and older age groups (see Norris, 2001). The second is the difference in political interest and engagement, as older cohorts, male voters, and more affluent social strata are more interested in the election and thus more attentive to its online development (Ifop, 2006). Voters who claimed to be actively participating in the campaign were significantly more likely to name the internet as their favorite avenue for political information than those who declared to be little engaged (Vedel, 2007, 4). Conversely, 61% of French internet users claimed to be interested in politics, while only 48% of the general population did (TNS Sofres, 2007, 26). One study found that interest in politics was among the strongest predictors of visiting both political and candidate websites (Vedel and Koc Michalska, 2007).

During the election, French internet users engaged in a variety of online activities, the most popular of which were responding to and consulting opinion polls (48 and 34%, respectively), reading commentary on forums and blogs (26%), forwarding links and documents to friends and acquaintances (26%), researching candidates' and parties' issue proposals (24%), and searching for humorous contents on the election (23%). Moreover, 23% of French internet users subscribed to a political newsletter, 18% viewed online videos on the campaign, 7% actively took part to political discussions on the internet, and 3% maintained their own weblog on the election (TNS Sofres, 2007, 20). Not surprisingly, web users relied mostly on traditional media's internet outlets to acquire information on the campaign: 63% claimed to log on newspapers' sites and 34% preferred television channels' sites, while generalist information portals, most of which relay contents from traditional media and press agencies, were favored by 50% of internet users. Parties' websites and candidates' personal sites were quite distanced, at 20 and 19%, respectively. Whereas most voters rationally trust news organizations more than political actors for their information on the campaign, parties' and candidates' sites remain more popular than, for example, citizen blogs and discussion forums (Ifop, 2006, 17). Moreover, visitors of parties' and candidates' sites are not solely motivated by the desire to acquire information on the election, but also by the opportunity to engage in the campaign.

For their part, political parties and candidates made unprecedented efforts to tap into this potential for online citizen engagement. One year from the election, both major parties, the Union pour un Mouvement Populaire (UMP) and the Parti Socialiste (PS), launched large-scale membership drives that made heavy use of the ICTs. While the UMP claimed to have recruited about 30,000 new members online between September 2005 and July 2006, the PS, which introduced a discounted 20-euro membership fee for new adherents, reported as many as 68,049 new subscriptions online (Astor, 2006). Moreover, PS Presidential candidate Ségolène Royal based a significant part of her campaign on the internet, which was instrumental in delivering her the nomination against the PS establishment in the primary elections held on 16 November 2006 among party members, many of whom had adhered online in order to be able to nominate Royal (Dolez and Laurent, 2007). During the campaign, the PS candidate asked her supporters to contribute to her platform with ideas and proposals, encouraged the creation of a geographically located network of supporter blogs (named 'Ségoland'), and had campaign staffers chat with site users on a daily basis. Media coverage of the race devoted large space to the internet, many political videos on peer-to-peer sites such as Dailymotion and YouTube achieved very high ratings, and a few Presidential candidates went as far as to create their digital headquarters on Second Life, the web-based virtual world. While some of these online endeavors by parties and candidates could be described as public relations efforts to achieve media coverage and an aura of 'glitziness,' other aspects of the French e-campaign provided citizens with substantial channels of expression and participation. In late March 2007, the Royal campaign claimed to have recruited about 250,000 online supporters, who had helped generate 150,000 policy proposals and 1,400 blogs (Desirs d'Avenir, 2007).

Top

Research Questions and Methodology

These premises led us to two main sets of research questions. The first regards the state and quality of online campaigning by parties and Presidential candidates in the French 2007 election. In light of the theoretical background previously summarized, we will thus analyze parties' and candidates' websites in order to evaluate the extent to which they informed and engaged their users. Furthermore, we will track the changes and developments in online activity that unfolded as the Presidential and then legislative campaigns progressed.

The second set of research questions explores the differences and similarities between the various French national political actors in adopting online tools. Three interesting comparisons can be made in this regard. First, between candidates and parties. Second, between top-tier and lower-tier parties and candidates. Third, between right-wing, center, and left-wing parties and candidates. Such assessments will help us better understand the factors that influence the patterns of adoption of ICTs by political actors.

To answer such questions, I performed a quantitative analysis of the characteristics of 22 websites: 10 of national parties and 12 of all the Presidential candidates who were on the ballot in the first round of the election.2 In order to track changes in websites as the Presidential and then the legislative elections neared, five separate measurements were made. The first two involved both parties' and Presidential candidates' sites and were performed, respectively, between 1 and 10 March 2007, and between 16 and 22 April 2007, thus ending on the day in which the first round of the Presidential election was held. The third measurement involved only the sites of the candidates who took part to the 6 May runoff, Ségolène Royal and Nicolas Sarkozy, and occurred between 23 April and 4 May. The fourth and fifth measurements only considered party sites and were completed, respectively, between 4 and 10 June, the day in which the first round of the legislative elections took place, and from 11 to 17 June, when the second round occurred.3

Each site was analyzed on the basis of a coding scheme that featured 81 variables, which were defined according to the study's theoretical background and building on previous international research. In particular, Gibson and Ward's (2000) seminal proposal constituted a key reference due to its clear characterization of five main functions of parties' websites: information provision, campaigning, resource generation, networking, and participation. Gibson and Ward also suggest quantifying 'site delivery,' which comprises both the 'glitziness' of its stylistic features and some measures of accessibility, navigability, responsiveness, visibility, and freshness. From these general categories, a set of individual variables can be derived to measure the various ways in which party functions and site sophistication are embodied in parties' websites. Other significant methodological contributions came from Norris (2003) and Farmer and Fender (2005). Following their research, each website was assigned a score for every variable (most often 0 for the absence and 1 for the presence of a characteristic). Subsequently, these values were aggregated into additive indexes that summarized the results for each website in each dimension of the analysis. The indexes were then divided by the maximum value that could be reached in each section in order to facilitate comparisons among the sites across all dimensions. Each index thus consists of a number between zero and one, which will be expressed as a percentage for easier reading.

The coding scheme devised for this study departs from Gibson and Ward's and Farmer and Fender's proposals in two ways. First, as the authors themselves acknowledge (Gibson and Ward, 2000, 307–308; Farmer and Fender, 2005, 50), this framework for analysis should be integrated by other scholars both theoretically, according to their research questions and perspectives, and empirically, following developments in online electioneering. Thus, several categories were added to account for applications that were not relevant when earlier studies were conducted, such as blogs, wikis, social networking tools, podcasts, and others. Second, while both Gibson and Ward and Farmer and Fender identify two main categories in which variables are divided (party functions and professional/technical functions, each in turn articulated into five sub-categories), the model employed here is composed of three macro-sections (information, participation, and professionalism), which are in turn divided into sub-sections. Thus, while professional and technical sophistication is maintained as an independent macro-area, the concept of 'party functions' is here divided into informational and participatory features. This distinction is necessary to account for the fact that, as Bimber and Davis wrote (2003, 48), the main goals of online campaigns are two-fold: 'to motivate supporters to take some kind of concrete action' (i.e., participation) and 'secondarily, to convert undecideds into supporters' (i.e., information). Similarly, Davis and Owen (2005) claim that opinion reinforcement and mobilization are the two main functions of parties' and candidates' websites. Such an approach was also reflected in the coding scheme adopted by Norris (2003, 29), where variables are classified along two dimensions: information transparency and communication interactivity, which is related to mobilizing resources and enabling citizens' expression. The relevance of such a distinction is confirmed by a recent study, which reveals that citizens who engage in internet politics pursue two distinctive goals — obtaining information and communicating with like-minded individuals — which are achieved through different online outlets (Kaye and Johnson, 2006).

Thus, the data for this research were collected, and will be presented here, in a scheme divided into three parts. The information macro-section accounts for 'pull' (user-initiated) and 'push' (party-initiated) information supply, and targeting of different groups of voters via dedicated tools. Participation entails online interactivity, resource mobilization, and decentralization of communication. Finally, professionalism is measured with respect to design and multimedia features, site accessibility, navigability, and frequency of updates.4

Top

Results

As the Presidential campaign unfolded, candidates continuously updated, revised, and renewed their websites. The data reported in Table 1 show a growth of five percentage points in the average quality index, an achievement not to be underestimated given that only 6 weeks passed between the March and April observations. The largest increases can be seen in the information and professionalism areas, while a much smaller improvement was detected for participation tools. In other words, candidates made their websites more informative and attractive as election day neared, but their efforts to increase online opportunities for citizen engagement were much smaller.


In the Presidential race, Sarkozy and Royal deserve a separate analysis, as their campaigns were protracted for 2 weeks longer than those of other candidates until the 11 May runoff. As Table 2 shows, the website of the PS candidate did not undergo any noticeable development in the final 2 months of the campaign, although a major restyling had occurred in the first weeks of 2007, as the candidate's goals changed from securing her party's nomination to winning the general election. Because Royal's pre-primary and primary campaign had already skillfully and intensely employed internet tools, her online apparatus was well developed and ready for the general election several months before the vote. On the other hand, Sarkozy successfully played catch-up with his opponent by adding a significant amount of functions to his website in the last 10 weeks of the campaign. Thus, his site's quality index increased by a remarkable 16 points from the first to the last survey. Both candidates engaged in a healthy process of reciprocal emulation, as the innovations that appeared on one contender's website were quickly adopted by the other.


Although Sarkozy updated his website to rival and eventually overtake Royal's in terms of the overall quality index, the two internet campaigns presented clearly distinctive profiles. Royal maintained a significant edge in the use of online tools for political participation, while Sarkozy's site was expanded especially in terms of information supply and professionalism.

Parties' websites show a more complex pattern, from the heated Presidential contest to the less intense legislative election that followed, where, as is often the case in the French semi-presidential system, the President-elect takes advantage of the 'honeymoon' with voters to bring his party to victory. Thus, as Table 3 shows, parties' online activities peaked in the week before the first Presidential round and then declined during the parliamentary election. While legislative campaigns are mainly conducted within electoral districts by individual candidates, parties clearly play a role in these elections, not necessarily smaller than in the Presidential contest, where the focus is more on national candidates than their parties. The five-point reduction in the overall quality index from April to June can be interpreted as a consequence of the low competitiveness of the legislative campaign, where little doubt existed as to whether UMP's parliamentary majority would be confirmed, to the point that even PS candidates by and large ran on the argument that, while defeat was certain, it would be important not to lose too many seats in order to keep a viable parliamentary opposition in place to limit Sarkozy's leeway.


Thus, while parties were strongly involved in the Presidential campaign, working in parallel with the candidates to inform, engage, and mobilize voters online, they subsequently abandoned a significant part of the internet tools they had created on the road to the Élysée. Quite remarkably, however, no net change or even positive changes were observed in the information and professionalism sections, but it was in the participation area that parties significantly downsized their efforts. Indeed, most participation tools require not just one-off software engineering, but continuous monitoring and care by party staffers. Incoming messages, photos, or videos from supporters must be scrutinized, filtered, and responded to. Activist and social networking tools, even though they are based on users' spontaneous initiatives and networking practices, need fine-tuning and work best when the campaign's presence on them is clearly felt. Blog postings, fundraising drives, calls for volunteers, petitions to be signed online, all require direct initiative from party staffers. As a consequence, while internet participation tools can offer parties and candidates very large rewards, they also entail some costs in terms of staff work and dedication. Apparently, parties perceived such efforts to be worthwhile during the Presidential campaign, but not in the legislative battle, where resources may be more soundly spent on local races than on the national stage.

The longitudinal, aggregate data we have presented so far do not reveal stark differences between parties' and candidates' online efforts. Both hover around 50% in the quality index, which is the result of greater performances in information and professionalism than in participation. A more detailed picture can be obtained by confronting candidates' and parties' sites in all macro-sections and sub-sections, as Table 4 does.


The information macro-section does not highlight stark differences between parties and candidates, although the former engage in voter targeting to a significantly larger degree. Website features that were classified in the professionalism macro-section also seem to be quite uniformly adopted, as is evidenced by the small coefficients of variation. It is the participation tools that display the largest variations, both in general, as candidates distance parties by five percentage points in the total index score, and specifically with respect to the section's three sub-components, where differences between parties and candidates are remarkable. Parties outperform candidates in resource mobilization, while candidates have a clear edge in both online interactivity and, especially, communication decentralization. Indeed, such a discrepancy can be explained by the different incentives and constraints that influence parties and candidates. Being permanent structures, parties need to mobilize resources not only during campaigns, but also outside of them, therefore they usually are better equipped and expedite at doing so than candidates, whose organizations, on the other hand, are much more short-lived. On the other hand, most parties are entrenched, change-averse organizations, less keen on relinquishing message control to their internet users, while some candidates (and their staffs) can be more dynamic and innovative, especially when the politician is a relative outsider from major party organizations. This can explain why candidates tend to embrace online discussion and decentralized communication more than parties do.

Although the aggregate data do not suggest dramatic dissimilarities between candidates and parties, many individual candidates adopted remarkably different internet strategies than those of their respective parties. A regression equation between the overall quality scores of the websites of the six candidates who were supported by a major party and those of their respective parties showed a statistically insignificant (P=0.359) correlation coefficient of 0.460, which resulted in an adjusted R2 as small as 0.015. Thus, candidates sometimes parted company from their parties in terms of internet use, as becomes evident in Table 5, where the four main parties' sites are compared with their respective Presidential candidates' sites.


The smallest gap in overall quality can be observed between UMP and its candidate, Sarkozy, although the similar indexes are the results of markedly different scores in participation, professionalism and, especially, information. The second smallest gap, already as large as 11 points, can be found between PS and Royal, with a conspicuous 17-point difference in participation. Finally, the distances between Bayrou, Le Pen and their respective parties are even greater. The former clearly invested in his personal site more than his party did,5 while the latter underplayed online tools compared to the FN.

A recurring theme in the literature on internet politics is the relationship between candidates' and parties' size and resources and the quality of their websites. As previously specified, normalization theory, which postulates a strong link between these variables, has been tested in different countries — usually employing vote share as a proxy for party size, as we did in the present work — with mixed results. Table 6 shows that in the French 2007 campaign the association between electoral viability and website quality was not negligible for both parties and candidates.


Electoral strength thus seems to be significantly related to parties' and candidates' adoption of ICTs. The differences, however, are smaller among candidates than among parties. This impression is confirmed by the results, reported in Table 7, of regression equations that tested the relationship between vote shares and the indexes of website overall quality. Parties and candidates were analyzed first separately, then combined. That the party landscape is more 'normalized' than the candidate pool cannot seemingly be accounted for by financial disparities alone. Indeed, adding financial data to the equations shown in Table 7 did not result in statistically significant coefficients and did not improve, rather in some cases decreased, the model's overall goodness of fit.6 The cause for the imbalances may not be related to resource availability per se, but to individual strategic decision-making on how to deploy them. Larger parties and stronger candidates might simply assess that their potential electorates are larger and therefore establish strong communication efforts in all available media outlets to cater to all their electoral segments, while smaller forces, needing to grow from modest voter bases, probably find it wiser to invest in traditional channels, such as mass media and grassroots structures, which still reach wider audiences. A speculative explanation for our finding of higher imbalances between parties than between candidates can be identified in the different availability of time to these two types of political actors. While parties' websites are built and maintained for extended periods and long-term objectives, candidates' websites by and large exist only in the relatively short time span of an election. Thus, even assuming that resource disparities have similar influences on website quality, they have more time to affect parties' than candidates' websites. While at the beginning of the campaign all candidates have to build their sites from the ground up, parties can improve on their existing ICT infrastructures. Thus, whereas all candidates have to do the same amount of work if they want to achieve similar website quality standards, some parties start from a better position than others and thus, even assuming similar efforts by all during the campaign, inequalities are bound to be more enduring among parties due to the persistence of earlier gaps. Another possibility is that candidate organizations, being more short-term and goal-oriented than party bureaucracies, adopt more innovative and dynamic strategies, which may include larger investments in the new media than traditional resource allocation would suggest.


Overall, the coefficients of correlation are higher than those that emerged from previous studies. For example, Cunha et al. (2003, 74) studied parties' websites in Greece, Italy, Portugal, and Spain and found coefficients ranging from 0.44 for Portugal to 0.35 for Greece. In a research on parties' websites in the 2006 Italian parliamentary elections, Vaccari (2008) found a correlation of 0.507. Instead, the quality of French parties' websites correlated more strongly (R=0.627) with their vote share. One possible explanation for this pattern might be the ability of the institutions of the Fifth Republic, especially the double-ballot electoral system adopted for legislative elections, to reduce the number of relevant parties (Sartori, 1976, 139–141; Knapp, 2004, 25–33; see also the proceedings of a recent symposium: Grofman and Lewis-Beck, 2005). Because they have few chances to gain seats in the national parliament and influence government policies, small parties in France have limited incentives to campaign vigorously, which is reflected in their weak online efforts. Electoral strength appears to be related to all dimensions of website quality in a similar way, as can be seen in Table 8, which shows the results of three regression analyses that estimate the relationship between parties' and candidates' vote shares and their websites' indexes of information, participation, and professionalism. Coefficients are of similar magnitude, although information has the strongest correlation with electoral strength and professionalism has the weakest.


Our final research question deals with the relationship between ideology and website quality. As Table 9 shows, whether candidates and parties are located on the right, center, or left of the political spectrum clearly makes a difference in terms of their internet investments. Centrist candidates and parties are the least attentive to ICTs, with significantly lower scores especially in information. Leftist parties offer richer websites than those aligned on the right, but the gap is negligible for candidates.


A closer analysis, however, reveals that the most interesting differences between right-wing and left-wing parties' and candidates' internet operations lie not in their sites' overall quality, but in the sub-section scores. Right-wing candidates and parties both register higher information indexes than their left-wing counterparts, which, on the other hand, report significantly stronger performances in participatory uses of the internet. While conservative political actors see the internet mostly as a tool to disseminate information, progressive candidates and parties are comparatively more eager to leverage on the dialogic, participatory, and organizational potential of the new technologies. Such a diversity in styles of ICT adoption is likely rooted in different types of relationships with supporters and voters, which tend to be more top-down for conservative parties and more bottom-up for progressive forces, as in the case of the participatory debates organized by Ségolène Royal to help draft her platform. Indeed, such differences in the strategic approach to voters appear to be highly rational, as leftist voters tend to have significantly wider participatory repertoires than centre and right-wing voters, who for the most part limit themselves to institutional acts such as voting (Muxel, 2006, 5). Because their target voters are more eager to become involved in the campaign, progressive parties and candidates tend to employ more engaging online tools, while conservative actors offer their internet users large amounts of information that they hope will strengthen their convictions and cement their voting intentions.

Top

Conclusions

There is little doubt that the 2007 elections have marked a turning point in French internet politics. If one looks at the overall quality scores for the websites of UMP and PS, as well as Sarkozy and Royal (see Tables 2 and 5), it is no surprise that the word Netcampagne became so popular in the French media and among the public during the election. Moreover, the successful online membership drives by UMP and, especially, PS, and the remarkable amount of issue contributions that Royal received via the internet signaled to all political actors that online tools have reached the point where they can deliver valuable real-world political resources. The intense and spirited Presidential campaign provided an incentive for both parties and candidates to improve their ICT infrastructures, although parties quickly scaled them back again during the quieter legislative campaign 2 months later. In other words, French parties have not yet fully embraced the internet as a medium enabling continuous and dialogic communication, and thus tend to allocate more technological resources and staff to their online operations when they are in full campaign mode than in less competitive phases.

At their highest point, both parties and candidates averaged a quality index that floated around the 50% mark, with greater scores in information and professionalism, lower in participation. Unless one disregards minor candidates and parties to focus only on UMP, PS, and their nominees, the process of integration of the new media into the arsenal of French contemporary campaigning is still far from complete, especially with respect to online user engagement. There is a large amount of variation in the way different political actors take advantage of ICTs. In April, while PS scored as high as 76%, both UDF and MPF did not go beyond 39%; while Sarkozy's and Royal's websites achieved 66% and 65% respectively, Arlette Laguillier's scored as little as 29%.

Asymmetries in the internet strategies of parties and candidates seem to be mainly related to two variables: electoral viability and ideology. Especially among parties, there is a clear difference between major and minor forces in the willingness and ability to fully experiment with ICTs. Such a gap is routinely found in Western European countries, but the data presented here suggest that it was larger than average in the French 2007 election. Two, not necessarily mutually exclusive, explanations can be advanced. First, continuous technological developments — most recently, the increasing popularity of web video and social networking tools — require increasingly larger investments than in the past to fully harness the political potential of the internet, thus making it more difficult, or less worthwhile, for resource-poor organizations to keep up with major forces. Second, as mentioned above, the French party system features a very deep divide between major and minor parties due to electoral and institutional constraints, which discourage fringe parties to campaign vigorously on the national stage as they have very few chances to achieve relevance.

Ideology also turned out to be a significant source of variation, as progressive parties were found to invest proportionally more on participation, while conservative forces focused especially on information provision. Distinctive styles of campaigning seem to be emerging online, whereby the right males the most of the internet's potential for disseminating information to various audiences in different formats, including rich audiovisual contents, while the left is more inclined to (cautiously) embrace a participatory philosophy that stimulates and channels user engagement. If such a trend is confirmed by future, comparative research, our understanding of internet campaigning might be enriched by the awareness that there is not one single model of online electioneering, but different political forces have diverse goals and incentives when they campaign in the new media, and thus develop distinctive communication styles in order to meet the needs and inclinations of their target voters. Thus, future research should also probe whether those differentiated efforts respond to, and satisfy, distinctive needs and habits among progressive vs conservative voters and internet users. As a highly selective and adaptable medium, the internet can indeed be expected to be more targeted and user-tailored than the traditional tools of mass-mediated political communication. Addressing these questions in future comparative studies would surely help us learn more about contemporary campaigning, not only in cyberspace but also in the offline realm.

Top

Notes

1 Data from Eurostat, see http://www.ec.europa.eu/eurostat.

2 The first round of the Presidential elections was held on 22 April, with 12 candidates on the ballot. The runoff between socialist Ségolène Royal and Gaullist Nicolas Sarkozy, who went on to win by a 53-47 margin, was held on 6 May. The first round of the legislative elections was held on 10 June and the runoff took place on 17 June. Sarkozy's party, UMP, obtained 313 parliamentary seats out of 577, while the PS had 186.

3 In addition, some parties' and candidates' sites were not considered in all observations. The sites of Presidential candidates Frédéric Nihous and Gérard Schivardi were not analyzed in the first measurement, but were included in the second after they met the signature threshold required to be on the ballot. The sites of newly formed parties Mouvement Démocrate and Nouveau Centre, both of which were born from the UDF after the Presidential election, were only considered in the fifth measurement. Appendix A lists all candidates' and parties' websites and their URLs.

4 The full coding scheme can be found in Appendix B.

5 At the time of the Presidential election, Bayrou was the leader of UDF. Afterwards, he left UDF and founded the Mouvement Démocrate (MoDem).

6 The results of such equations are not presented here because, while expenditures by Presidential candidates are publicly available at the time of this writing, only data referring to the 2004 general budgets are available for parties, which, to make matters worse, forced us to exclude MoDem and Nouveau Center from the analysis as they were formed in 2007. In order to conduct a reliable study of these relationships, 2007 financial data for parties will have to be included.

Top

References

  1. Astor, P. (2006) 'Présidentielle: Les nouveaux adhérents en ligne, une force pour les partis politiques?', 4 August, http://www.zdnet.fr/actualites/internet/0,39020774,39362472,00.htm.
  2. Bentivegna, S. (2006) 'Rethinking politics in the world of ICTs', European Journal of Communication, 21(3): 331–343. | Article |
  3. Bimber, B. and Davis, R. (2003) Campaigning on Line: The Internet in US Elections, Oxford, New York: Oxford University Press.
  4. CEVIPOF (2007a) 'Le Baromètre Politique Français, 4ème vague', CEVIPOF-Ministère de l'Intérieur, http://www.cevipof.msh-paris.fr.
  5. CEVIPOF (2007b) 'Le Panel Electoral Français 2007: Enquête post électorale présidentielle 2007', CEVIPOF-Ministère de l'Intérieur, http://www.cevipof.msh-paris.fr.
  6. Cunha, C., Martìn, I., Newell, J. and Ramiro, L. (2003) 'Southern European Parties and Party Systems, and the New ICTs', in R. Gibson, P. Nixon and S. Ward (eds.) Political Parties and the Internet: Net Gain?, London, New York: Routledge, pp. 70–97.
  7. Dalton, R. and Wattenberg, M. (eds.) (2000) Parties without Partisans: Political Change in Advanced Industrial Democracies, Oxford: Oxford University Press.
  8. Davis, R. and Owen, D. (2005) 'The United States: The Internet and the 2004 Presidential Election Campaign', Parties and Election Campaigning Online, Proceedings of the Oxford Internet Institute International Workshop, 20–21 October 2005; Oxford, UK.
  9. Desirs d'Avenir (2007) 'Point d 'étape sur la Net-campagne de Ségolène Royal', 21 March, http://www.desirsdavenir.org/pdf/point_d_etape.pdf.
  10. Dolez, B. and Laurent, A. (2007) 'Une primaire à la française: La designation de Ségolène Royal par le parti socialiste', Revue Française de Science Politique, 57(2): 133–161.
  11. Farmer, R. and Fender, R. (2005) 'E-parties: democratic and Republican State parties in 2000', Party Politics, 11(1): 47–58. | Article |
  12. Gibson, R. and Ward, S. (2000) 'A proposed methodology for studying the function and effectiveness of party and candidate web sites', Social Science Computer Review, 18(3): 301–319.
  13. Grofman, B. and Lewis-Beck, M. (2005) 'Symposium introduction: elections under the French double ballot system', French Politics, 3(2): 93–97. | Article |
  14. Ifop (2006) 'L'observatoire 2006-2007 de la netcampagne présidentielle', http://www.ifop2007.fr.
  15. Kaye, B. and Johnson, T. (2006) 'The Age of Reasons: Motives for Using Different Components of the Internet for Political Information', in A. Williams and J. Tedesco (eds.) The Internet Election: Perspectives on the Web in Campaign 2004, Lanham: Rowman & Littlefield, pp. 147–167.
  16. Knapp, A. (2004) Parties and the Party System in France: A Disconnected Democracy?, Basingstoke: Palgrave Macmillan.
  17. Margolis, M. and Resnick, D. (2000) Politics as Usual: The Cyberspace 'Revolution', London: Sage Publications.
  18. Margolis, M., Resnick, D. and Levy, J. (2003) 'Major Parties Dominate, Minor Parties Struggle: US Elections and the Internet', in R. Gibson, P. Nixon and S. Ward (eds.) Political Parties and the Internet: Net Gain?, London, New York: Routledge, pp. 53–69.
  19. Mediametrie (2007) 'L'audience de l'Internet en France, Avril 2007', http://www.mediametrie.fr.
  20. Muxel, A. (2006) 'Les Français et la participation politique', Le Baromètre Politique Français 2006–2007, vague 2, CEVIPOF-Ministère de l'Intérieur, http://www.cevipof.msh-paris.fr.
  21. Nixon, P., Ward, S. and Gibson, R. (2003) 'Conclusions: The Net Change', in R. Gibson, P. Nixon and S. Ward (eds.) Political Parties and the Internet: Net Gain?, London, New York: Routledge, pp. 234–243.
  22. Norris, P. (2001) Digital Divide: Civic Engagement, Information Poverty, and the Internet Worldwide, Cambridge, MA: Cambridge University Press.
  23. Norris, P. (2003) 'Preaching to the converted? Pluralism, participation and party websites', Party Politics, 9(1): 21–45. | Article |
  24. Sartori, G. (1976) Parties and Party Systems: A Framework for Analysis, Cambridge, MA: Cambridge University Press.
  25. Sauger, N. (2002) 'Les partis sur le Net: Première approche des pratiques virtuelles des partis politiques français', in V. Serfaty (ed.) L'Internet en politique, des Etats-Unis à l'Europe, Strasbourg: Presses Universitaires de Strasbourg, pp. 179–195.
  26. Schweitzer, E. (2005) 'Election campaigning online: German party websites in the 2002 national elections', European Journal of Communication, 20(3): 327–351. | Article |
  27. Stromer-Galley, J. (2000) 'On-line interaction and why candidates avoid it', Journal of Communication, 50(4): 111–132. | Article |
  28. TNS Sofres (2007) 'Les internautes et la politique', http://www.tns-sofres.com.
  29. Vaccari, C. (2008) 'Research Note: Italian Parties' Websites in the 2006 Elections', European Journal of Communication, 23(1): 69–77. | Article |
  30. Vedel, T. (2007) 'Les électeurs et la campagne électorale: Une vision désenchantée des candidats?', Le Panel Électoral Français 2007, vague 2, CEVIPOF-Ministère de l'Intérieur, http://www.cevipof.msh-paris.fr.
  31. Vedel, T. and Koc Michalska, K. (2007) 'Political participation and the internet: evidence from the 2007 French presidential election', Proceedings of the ICEG 3rd International Conference on e-Government; 27–28 September 2007; Montreal, Canada.
  32. Villalba, B. (2003) 'Moving towards an Evolution in Political Mediation? French Political Parties and the New ICTs', in R. Gibson, P. Nixon and S. Ward (eds.) Political Parties and the Internet: Net Gain?, London, New York: Routledge, pp. 120–138.
  33. Ward, S., Gibson, R. and Nixon, P. (2003) 'Parties and the Internet: An Overview', in R. Gibson, P. Nixon and S. Ward (eds.) Political Parties and the Internet: Net Gain?, London, New York: Routledge, pp. 11–38.
  34. Ward, S. and Vedel, T. (2006) 'Introduction: the potential of the internet revisited', Parliamentary Affairs, 59(2): 210–225. | Article |
  35. Williams, A. and Tedesco, J. (eds.) (2006) The Internet Election: Perspectives on the Web in Campaign 2004, Lanham: Rowman & Littlefield.
Top

Appendices

Appendix A

List of parties' and candidates' websites analyzed

Parties
 

Front National (FN) — http://www.frontnational.com

Les Verts — http://www.lesverts.fr

Mouvement Démocrate (MoDem) — http://www.bayrou.fr

Mouvement pour la France (MPF) — http://www.mpf-villiers.com

Nouveau Centre (NC) — http://www.le-nouveaucentre.org

Parti Communiste Français (PCF) — http://www.pcf.fr

Parti Radical de Gauche (PRG) — http://www.planeteradicale.org

Parti Socialiste (PS) — http://www.parti-socialiste.fr

Union pour la Démocratie Française (UDF) — http://www.udf.org

Union pour un Mouvement Populaire (UMP) — http://www.u-m-p.org

Presidential candidates
 

François Bayrou (UDF, then MoDem) — http://www.bayrou.fr

Olivier Besancenot (Ligue communiste révolutionnaire) — http://www.besancenot2007.org

José Bové (anti-liberal collectives) — http://www.unisavecbove.org

Marie-Georges Buffet (PCF) — http://www.mgbuffet.org

Arlette Laguiller (Lutte ouvrière) — http://www.arlette-laguiller.org

Jean-Marie Le Pen (FN) — http://www.lepen2007.fr

Frédéric Nihous (Parti Chasse, pêche, nature, traditions) — http://www.nihous2007.fr

Ségolène Royal (PS) — http://www.desirsdavenir.org

Nicolas Sarkozy (UMP) — http://www.sarkozy.fr

Gérard Schivardi (Parti des Travailleurs) — http://www.schivardi2007.com

Philippe de Villiers (Mouvement pour la France) — http://www.pourlafrance.fr

Dominique Voynet (Les Verts) — http://www.dominiquevoynet.net

Appendix B

Coding sheet for website characteristics

1. Information tools (maximum score=32 for parties, 30 for candidates)

1a.Information supply (maximum score=22 for parties, 20 for candidates): History, 'about us' — Leader/candidate's profile and biography — Candidates' profile and biography (only for party sites) — Party/campaign structure and organization — Contacts, phone numbers, e-mail addresses — Endorsements by groups, associations, movements — Party values and ideology — Party issue platform — Speeches and statements by party officials — Events calendar — Press clippings — Information on electoral law and voting procedures — Comparisons with opponents' record and platform — Data, statistics, dossiers on public policies — External links — Dedicated election website (only for party sites) — Downloadable materials from the communication campaign — Frequently Asked Questions on the party/candidate — Other modalities — Average number of news items published daily (0 items=0 points; 1 or 2=1; 3 or 4=2; from 5 to 7=3; more than 7=4)

1b.Information in push mode (maximum score=4): Newsletter subscription — SMS and PDA updates — Podcasting — XML, RSS feeds — Other modalities

1c.Targeting (maximum score=6): Specific pages for interest groups (e.g. business, labour, teachers) — Specific pages for issue groups (e.g. environmentalists) — Specific pages for identity groups (e.g. women, gays, minorities) — Specific pages for age groups (e.g. young, seniors) — Specific pages for voters in different geographic areas — Newsletters tailored to users' subscription information — Other modalities

2. Participation tools (maximum score=30)

2a.Online interactivity (maximum score=13): Online polls — Discussion forums, chat rooms — Chats and forums with candidates and party or campaign officials — Solicitation of questions to candidate/party/campaign officials via web or e-mail — Publication of questions sent and answers to them — Blog by the candidate or party — Users' comments allowed on the blog — Wiki and other forms of collective writing — Guestbook — Publication of brief messages by users (via online form or SMS) — Other modalities — Time required to answer an e-mail question on issues (one business day=3 points; from 2 to 4 business days=2; one business week and more=1; no answer=0)

2b.Resource mobilization (maximum score=9): Site area dedicated to volunteers and campaign engagement — Online subscription to the party (only web-based mode considered) — Online volunteer sign-up — Online fundraising (only web-based mode considered) — Online gadget store — 'Send this page to a friend' — Other modalities — Time required to answer a volunteer offer by e-mail (one business day=3 points; from 2 to 4 business days=2; one business week and more=1; no answer=0)

2c.Communication decentralization (maximum score=8): Social networking tools — Volunteer recruitment tools — Media contacting tools — Offline distribution of online-gathered materials (e.g. leaflet printing) — Tools for users to build sites linked to party/candidate site — Tools for users to build blogs linked to party/candidate site/blog — Online distribution of materials (e.g. electronic cards, e-mail forwarding, banners) — Possibility to send and publish user-produced audiovisual materials — Other modalities

3. Professionalism (maximum score=23)

3a.Design and multimedia features (maximum score=10): Animated introduction — Interstitial (splash) page before the home page — Photos and graphics (besides the header) — Background music — Audio clips — Video clips — Audio/video live streaming — Moving objects within the page — Online games — Animations, cartoons, comics

3b.Accessibility (maximum score=4): Access for disabled people (logo shown) — Accessibility with different browsers (W3C consortium logo shown) — Translation in other languages — Privacy statement

3c.Navigability (maximum score=5): Navigation bar in all pages — 'Back' or 'home page' link in all internal pages — Internal search engine — Site map — Navigation guide

3d.Update frequency (maximum score=4): Date when last update occurred (1 month ago=0 points; 2 weeks ago=1; last week=2; 1–2 days ago=3; within the last 24 hours=4)

Total maximum score=85 for parties, 83 for candidates.