Référence : 1

Foreword

Digital tools have, over the years, profoundly transformed electoral campaigns. Social networks change the way voters "follow" the campaign (polls, initiatives and political "moves" of the candidates, its adventures). They allow the most concerned of them to take an active part in the campaign: by relaying the themes, the arguments, the videos of their candidate.

The campaign teams, for their part, are implementing real digital strategies to encourage, activate and coordinate the commitment of these digital activists of a new kind, most of whom are not members of political parties.

In this sense, digital technology enriches and renews the " election action directory " with new ways for citizens to get involved, and new ways for candidates and their teams to " campaigning " for candidates and their teams.

Despite a profusion of barometers that scrutinize the digital presence of candidates and the mobilization of their supporters, the question of the impact of social networks and new forms of "digital activism" on the formation of electoral choices, and more broadly on the "manufacture of opinion" remains largely open.

Référence : 2

Contents

• What digital uses during the election period?
• 2022: A profusion of barometers to scrutinize the engagement and mobilization of voters on social networks
• Social sciences point to an overestimation of the contribution of social networks to opinion formation
• Instant messengers, the invisible face of digital campaigns
• New research, new questions about the political uses of digital technology

Références : 3 4 5

What digital uses during the election period?

The main source of knowledge about the digital practices of voters remains theFrench electoral survey (ENEF), conducted by Ipsos on behalf of Cevipof, Le Monde and the Jean Jaurès Foundation.

Based on a panel of 10,000 people, interviewed eleven times between April 2021 and May 2022, the ENEF combines questions on opinions and voting intentions with questions on the means of informing voters and on their modes of engagement. It "makes it possible to study minority practices (for example, liking a candidate) and to analyze the profile of certain groups, operations that would have no statistical significance with a standard sample of 1,000 individuals, even if perfectly representative.

To date, the Cevipof has published only fragmentary data on the information practices of voters during the 2022 presidential election.

In the first place, they were questioned about "the 15% who said they had". The second most popular means of obtaining information, just after political programs on television, but ahead of the written press and the programs of the candidates received in the mailboxes.

Based on the French Electoral Survey, Thierry Vedel distinguished three types of informational practices in politics:

As regards the social networks, Thierry Vedel observed that "concluded

Référence : 6

The survey found that only 7% of the 11,000 respondents cited "as" and 19% cited "as" as one of the "main ways to do it.

December 2021, social networks, forums and blogs "first means of informing oneself on political news

In Aprildifferent ways of getting information about the candidates' programs, which one did you prefer?", "looked for information on the Internet

2017: three types of informational practices in politics2017 campaign,

• About 15% of voters have intensive and multi-modal information practices, i.e. they use several sources of information by combining traditional and online media: regular viewing of television news, sharing of information on social networks, viewing of campaign clips, reading of campaign pledges, etc. During the campaign, this group increased their online political activity, but moderately so since it was already high to begin with.
• Other voters (about 45% of the population) have more fragmented and sporadic information practices. They generally combine two media (television and print media for older voters, television and social networks for younger voters). It is in this group that online political activity increased the most during the campaign.
• Finally, a third group (about 40% of the population) has extremely limited informational practices in political matters, which often boil down to watching television. This group, although strongly present on social networks, uses the Internet relatively little to follow the elections.

2017: It is mainly the most politicized and most educated voters who are the most active while about two thirds of voters have a profile or an account on social networks, only a minority use them to exchange or express political opinions. It is mainly the most politicized and educated voters who are the most active. Beyond the search for information, the Internet has not (yet) become a major space for political debate. This does not mean that it is without importance in the electoral competition. If "there is a distinction to be made", If there is a distinction to be made, "it does not lie in the sources but in the informational practices (...). The main factor that controlled these informational practices during the presidential campaign is the interest in politics: the higher the interest, the more diversified and stronger the informational practices (...). It is not so much the Internet that leads to politics, but rather the intensity of politicization that leads to recourse to the Internet.

2022: A profusion of barometers to scrutinize voter engagement on social networks

It seemed to be a given for many observers - and for the media - that social networks were going to be an essential, if not decisive, stake in the presidential campaign: both a playground for the candidates and their supporters, and an observation ground that would allow them to follow, or even anticipate, changes in opinion and campaign dynamics.

Several factors explain this focus of attention on social networks: A profusion of barometers and observatories have been implemented by specialized firms to measure the visibility of candidates on social networks and to scrutinize the mobilization of their relays and supporters: number of posts mentioning each candidate, percentage of subscribers gained by the candidates, average number of interactions per publication on one or another of the social networks, Facebook publications that have generated the most reactions, videos related to the presidential election the most liked on YouTube or TikTok, the most shared tweets, the most liked Instagram post etc.

• First, the idea took hold, at the end of 2021, in a still uncertain health context, which limited large gatherings (no candidate wishing to see one of his meetings turn into a cluster) that the outcome of the campaign would be played, in part, on social networks.
• The advent of new platforms (TikTok and Switch) frequented mainly by a young adult audience whose abstention rate is often high. If the digital campaign had been deployed, in 2017, mainly on YouTube, Facebook and Twitter, the 2022 campaign saw the emergence of new digital arenas, Instagram, TikTok and Switch, and, with these platforms, the deployment of new formats: the participatory interviews of Twitch, the short formats of TikTok.
• The irruption in the campaign of a new generation of influencers, who displayed their willingness to reconcile youth with politics, such as HugoDécrypte, and its 1.9 million subscribers on TikTok, 1.6 on Instagram, 2 million on YouTube, 214 000 on Twitch.

So many measures and metrics widely commented by the media:

• Twitch, Snapchat, TikTok: when digital technology tries to reconnect young people to politics?
• Which candidate did well on social networks in January?
• Which candidate made the most noise in December on social networks?
• Can Twitch reduce youth abstention?

Références : 7 8 9 10 11 12

Social sciences point to an overestimation of the contribution of social networks to opinion formation

Around the place of digital technology in electoral campaigns, a field of research has emerged, at the crossroads of political science, sociology, data sciences and information sciences.

Academics, invited by the media to comment throughout the campaign on the evolution of the number of subscribers to the Facebook or Twitter accounts of a particular candidate, or the differential mobilization of their supporters, showed great caution.

"The effects attributed to social networks in terms of influence on the outcome of elections are overestimated"

For Fabienne Greffet, a researcher in political science at the University of Lorraine, in February, on FranceBleu, "the effects that we attribute to social networks in terms of influence on the outcome of elections are overestimated. It is difficult to relate the number of people involved with what is seen on social networks, there is a whole part of staging and therefore it must be borne in mind that we can also be faced with dozens of strategies for seeking influence, without the groups mobilized represent hundreds or thousands of people. And precisely, one of the effects sought may be to make people believe that a candidate is very supported or very attacked on social networks, when in fact, the operation is led by a small group.

"When we look at the mapping of networks, we see that the most active digital communities are ideologically very homogeneous. In the end, there is little exchange between communities. It's more often a matter of exchanging content with people who think like us since generally, our relational network is rather made up of people who think like us."

Active citizens on the Internet more radical 

Five researchers, Marie Neihouser, Felix-Christopher von Nostitz, François Briatte, Giulia Sandri and Tristan Haute reminded us, in April 2022, in TheConversation that the electoral uses of social networks, although they have been developing since 2012, remain relatively minor."Comments following candidates' messages remain relatively few on both Twitter and Facebook (...) if we relate them to the number of registered voters, or even to the number of people registered on social networks in France (40 million monthly users of Facebook, 8 on Twitter, 22 on Instagram, 50 on YouTube). Above all, some candidates only get a few hundred comments - or even less. Even if we look at the number of likes, a practice that is less costly for Internet users than commenting, reactions to the candidates' messages remain relatively rare, especially when compared to their number of subscribers. "We know that politically active citizens on the Internet have different characteristics: they are more interested in politics, more educated, and younger than average.

Online publications primarily reach audiences already committed to the cause

"For several weeks and sometimes several months, the candidates have been occupying their territory of visibility as much as possible, that is to say, the entirety of social networks - or almost - and traditional media. But it is very difficult to measure the impact that the various forms of digital communication can have on electoral behavior," observed, for her part, Stéphanie Wojcik (University of Paris-Créteil), interviewed by Usbek & Rica." It is often said, and rightly so, that online publications primarily reach audiences that are already 'captive', namely activists and supporters - in short, all those already committed to the cause."" For candidates, the whole point is to broaden the usual circle of their audience. Of course, it seems obvious that if you acculturate correctly to a specific device like TikTok, on which you produce short videos with a rather humorous or light tone, you will capitalize on the audience of this network (...). But there can also be a contradiction between these imperatives and electoral political communication as such, which is not always satisfied with such short durations. Sometimes it can be downright counterproductive for a politician to try to go on a platform where they are not very comfortable."

For Jen Schradie, a researcher at SciencePo (author of"L'illusion de la démocratie numérique. Is the Internet right-wing?"), "social networks function as megaphones or projectors, which distort what is happening in reality in many ways, and the French presidential campaign has not escaped this" .

Social networks in the campaign as an open-air meeting

For Olivier Ertzscheid, Olivier Ertzscheid, lecturer in Information and Communication Sciences at the University of Nantes, author of the blog Affordance, it is the image of the meeting that appears " the most appropriate to arbitrate on the role and the place that social networks really play in an election".

"A political meeting is a reflection of a dynamic: the ability to have a more or less full house, the ability to disseminate elements of language and analysis that will then structure the campaign and opinion, the ability to convince and mobilize of course. Social networks are all this at the same time". More than 90% of the people who come to a meeting are already convinced and only a few curious or opponents come to complete the ranks. But the way in which one addresses a room composed of 90% of convinced people, and the way in which this room acts and reacts, is out of all proportion to the way in which one can address a room which would be hostile or in any case composed of at least as many convinced people as opponents. It is more or less the same thing on the generalist social networks".

Dominique Cardon on RadioFrancereminded newsrooms that they "must have a digital culture of investigation. They must be able to evaluate the prevalence of a theme in public opinion. And above all, not to consider that what is said on Twitter is a reflection of public opinion. This would be a serious mistake. Because only a very small part of the French population is on this social network. Often they are journalists, graduates and urban people. Not very representative of the rest of the population. The most commented subjects on Twitter are not necessarily those that interest the French".

Références : 13 14 15 16

Instant messengers, the invisible face of digital campaigns

Marie Turcaneditor-in-chief of the Numerama website, draws attention to " what we don't see... Behind what is public, Facebook, Twitter, you have another digital landscape. Today, many French people get their information through these alternative channels, which are loops and WhatsApp or Telegram messengers (...). These are not very visible phenomena, but they can count and have influence. I think we should not minimize the impact of these subgroups that can gather hundreds of thousands of Internet users.

A point of view shared, in Mediapartby Jen Schradie: "we are probably missing a massive phenomenon that remains a black box, namely the private loops on WhatsApp, Telegram, Signal or private groups on Facebook, while this is probably where part of the digital politics is played.

Olivier Ertzscheidfor his part, sees " in the whole of these invisible shares of contents (...) which can thus be neither observed, nor measured, nor counted (or in any case much more difficult) a space henceforth determining and structuring for the whole of the movements and the political or citizen mobilizations".

He points out, in this regard, "a paradox of visibility that can be expressed in the following way: the content thus 'published' can enjoy a very high visibility in the private groups where they are disseminated or taken up, but their public 'analytical' visibility, the one that allows us to observe and take into account their virality, is, on the other hand, almost nil. Olivier Ertzscheid recalls, in passing, that WhatsApp had been used extensively by supporters of candidate Jair Bolsonaro to get him elected during the 2018 presidential campaign.

The hypothesis of a tripartition of the political space 

Olivier Ertzscheid sees a "tripartition taking shape for all future debates and elections, in France as elsewhere.

• First and foremost, the mass media surface of the political agenda and partisan influences remains (TV and the press continue to play a determining and crucial role in the manufacture of opinion).
• In second place and in second tempo comes the partisan interpersonal social surface that can be read explicitly in the generalist social networks and in their own editorial logic (the Trending Topics of Twitter for example).
• Thirdly, there are the sub-media spaces of influence (the Dark Social), mainly made of private messaging, which today give the third tempo of the electoral and political agenda.

Références : 17 18 19 20 21

New research, new questions about the political uses of digital technology

Fabienne Greffet and Marie Neihouser have undertaken, in a call for papers on "the digitization of electoral action repertoires" , to retrace the debates that cross the community of researchers who are interested in the political uses of digital tools. "At the turn of the 2000s, the specialized English-language literature was strongly divided around one question: can the investment of digital spaces confer additional resources to political formations or candidates previously too small to access elective positions, as well as to the mass media?"

In the 2010s, research focused on the professionalization of digital campaigns, with the use of specialized personnel: digital communication consultants, web communication managers, community managers, and data scientists.

In parallel, a questioning "on the degree of initiative of the engaged citizens on the digital spaces" is born with, according to the authors, a tension between two types of analyses: those which defend the idea according to which "to make a digital campaign consists in perfecting pre-existing marketing techniques of targeting" and those which estimate, on the contrary, " that the online campaigns are partly transferred from the partisan organizations towards the engaged citizens".

In France, research focuses on the digital practices of the people most involved in parties. "In fact, what people who engage online for a candidate or a party in a possibly punctual way do is not well known, especially when it comes to people who are more distant from the organization, called "sympathizers", whose activities receive little attention from researchers. Moreover, the lively controversy launched by the essayist Evgueny Morozov, on slacktivism or "soft engagement", has contributed to devaluing digital engagement activities in comparison to field action, and to considering them as decoys that give a good conscience by creating the illusion of acting when they have no real impact on mobilization or on the results of an election.

In conclusion, Fabienne Greffet and Marie Neihouser propose three areas of research and reflection to future contributors:

• The resources and actors of the digitalization of electoral action repertoires. "While some research focuses on the professionalization of online campaigns and the consequent appearance of new actors (data scientists, etc.), the resources made available to these new personnel in order to digitize the repertoires of electoral action remain little studied. Do they use the same software, the same information systems? Do they promote digitalization "standards"? In addition, the question arises of understanding how roles are divided within campaign teams between "professionals" (including outside companies) and "amateurs", but also between organizations and sympathetic digital communities, or between actors located in the space of electoral competition and actors located geographically outside (especially abroad).
• The digitalization of electoral action repertoires in practice. "How are new media, such as social networks, invested and with what objectives? To what extent are the repertoires of electoral action the continuation of more traditional repertoires of action by other means? To what extent do we observe ruptures, relying for example on "computational propaganda" techniques or on collective digital actions such as astroturfing, i.e. the artificial creation of what appears to be a mass and spontaneous online movement?
• How do these techniques and skills fit into broader campaign repertoires, especially offline?"
• The consequences of the digitalization of electoral action repertoires. "In this third axis, the consequences of the digitalization of the repertoires of electoral action will be researched - both from the point of view of the organization and the power relations within the parties and the campaign teams, and in terms of activities and contents produced, as well as in terms of the evolution of the interest and the mobilization of citizens during electoral campaigns.

Références : 22 23 24