2021-4354-AJSS-SOC – 02 JUL 2021

1 The Relevance of Scientific Dissemination during the 2 Campaign: The Italian Virologist 3 Communication on Social Media 4 5 The scientist role has progressively gained an essential relevance along the 6 2020 pandemic. The virologists’ exposition turned out, in fact, as 7 fundamental for the public opinion both for the well informed and not aware 8 people about health, transmission, and, today, program. 9 This paper aims to set a first explorative investigation about the social 10 communication practices during the first three months of the vaccine 11 campaign addressed by Italian most established virologists on social media. 12 The arising of digital scenario and the resultant pervasive presence in our 13 daily life of web platforms, such as social media, has revolutionized the 14 nexus between science and society. More scholars argued about the 15 disintermediated current shape of science communication that connect 16 directly scientist and large publics, driving the sociological debate towards 17 the analysis of the current processes of sense-making construction. On this 18 assumption we aim to give answers to the research question referred about 19 how do the Italian scientists communicate and approach large publics on 20 social media. Therefore, the empirical part of this paper consists in a data 21 collection phase conducted on and . The collected data 22 have been analyzed by a content analysis oriented to identify the 23 contradictory or uniformity of disintermediated communication features of 24 the observed social media profiles in order to push and follow, during the 25 ongoing vaccine dosing program, a proactive reflection about the key role of 26 scientific dissemination. 27 28 Keywords: Communication of science, Scientists’ visibility, Social media 29 communication, Content Analysis, Topic Modelling 30 31 32 The Communication of Science inside and outside the Digital Environment 33 34 The relationship between science and society has become nowadays 35 increasingly relevant. Following Ancarani (1996) in fact, science, gradually, 36 has been faced with a variety of politically relevant social and economic issues 37 such as health, food, transport, communications, energy, innovation and so on., 38 meanwhile the public space got transformed focusing citizens as «the 39 depositary of the structures and processes of democracy as the power control, 40 the delegation of people’s will, public discussion and public opinion» 41 (Mazzoleni, 2004: 17). 42 Today, therefore, institutions representing science and the researchers 43 themselves cannot neglect communication in their daily work (Scamuzzi and 44 Tipaldo, 2015). 45 In particular, science’s public communication has an essential role to 46 define the relationship between researcher and citizens (Bucchi and Trench, 47 2014) and now, a scientist is socially appreciated if he manages to reduce the

1 2021-4354-AJSS-SOC – 02 JUL 2021

1 distance between these subjects. In this perspective is pointed the need for a 2 close relationship between science and society, between experts and the public, 3 in a process of involvement that allows us to grasp the urgent needs of 4 humanity» (Pellegrini, 2018: 33). 5 Beyond the different levels of analysis of communicative phenomena and 6 consistent models that have outlined the relationship between science, 7 scientists and the various types of audience (Jasanoff, 1995), scientific 8 communication studies agree about how the mediatic context is «a central 9 variable for the regulation of the processes of constructing meaning, able to 10 orient exchanges between the issuer and the recipient towards outcomes not at 11 all obvious, even with equal content of messages» (Scamuzzi and Tipaldo, 12 2015: 68). In light of this assumptions, nowadays Internet represents certainly 13 the mediatic environment that has succeeded in revolutionizing the 14 communication of science simultaneously to its evolution. Firstly, intended as a 15 canonical medium deputed to the information storage (read-write web), the 16 web changed along last 20 years transforming itself in an integrated 17 participative environment (people-centric web and participative web, Patel, 18 2013). 19 There are many web-based activities that scientists carry out daily, as for 20 example documents transmission, magazines editing, data sharing, articles 21 creation, publishing of conference proceedings and informal exchanges also via 22 i.e. videoconferences. In this way, Pellegrini and Saracino in fact argue about 23 “Science 2.0” as an «increasingly widespread practice among scientists of 24 publishing online experimental results, emerging theories, claims of 25 discoveries and drafts of articles that anyone can read and comment on» 26 (Pellegrini and Saracino, 2019: 76). 27 Furthermore, Science 2.0 cannot disregard a communication also through 28 Social Media sphere, by which is possible to create networks of collaborations 29 (think i.e. of Researchgate designed to allow relationships between researchers 30 related to any type of discipline), to spread news and fight against scientific 31 controversies: today science and society work together in fact also because 32 they allow citizens in taking stand on scientific issues: a not practicable way 33 until recently when was «an exclusive prerogative of the scientific community 34 and political decision-makers» (Bucchi, 2010: 141). 35 Before Internet affirmition, the mainstream media as radio and Tv were 36 the only promoters of communication of science to citizens although «by their 37 nature they point out the evident asymmetry between the scientific community 38 and the general public and the clear directionality of science communication» 39 (Scamuzzi and Tipaldo, 2015: 150). In the past, traditional media acted as an 40 intermediary between universities and the public sphere for example through 41 press releases while, with the advent of digital, information shifted in an open 42 access vocation, accessible to anyone is interested. In this way the online media 43 offer to scientists more communication opportunities in dealing directly with 44 the general public rather than having to rely on journalists as mediators (Peters 45 et all., 2014).

2 2021-4354-AJSS-SOC – 02 JUL 2021

1 The web, furthermore, connotes as an environment in which comes 2 possible a greater participation about scientific knowledge and, at the same 3 time, that can reveals traps related to scientific controversies or misinformation 4 in the way that «the web breaks that sequential order and the tightness of a 5 series of "filters" that previously distinguished the path of scientific results of 6 the researcher to the general public» (Bucchi, 2006: 72). 7 Although its definite limits, «the web can allow a faster and immediate 8 access to scientific information (possibility to access original papers, databases, 9 contacts of researchers) » (Scamuzzi and Tipaldo, 2015: 150) and, in addition, 10 social networks become useful tools to simplify the communication of science 11 and its results as well the understanding of how a research has been led. 12 Following Bucchi and Saracino (2021), it is relevant finally to point out 13 how the science communication concept for scientific argues has been recently 14 rethought also due to the increasing of public demand about science and 15 technology argues. It overcomes in fact the classical processes of mediated 16 communication shifting to direct interaction between sender and public, driving 17 towards a model pointed in 3 specific strategies (Pellegrini and Rubin 2019: 18 71-72): 19 20 1. The vertical one: featured by the direct dissemination of press release 21 and scientists’ statements in order to spread the research outcomes to a 22 general public. 23 2. The dialogic one: featured by discussion events between experts and 24 publics duly shaped on new scientific topics. 25 3. The participative one: the last strategy concern the direct involvement 26 of individuals in the research works in order to enrich the research 27 purposes and shape the right interesting topics of investigation. 28 29 According with this background, this paper aims to shed light on the 30 current relation between scientists and citizens in . Though a first 31 exploration oriented on the scientists’ social media presence, in fact, we will 32 try to comprehend how they set their communication strategies and styles 33 selecting as unit of analysis the cultural products of their posting activities. For 34 this reason, the empirical part of our work starts from a specific research 35 question: referring the topic of current campaign, how do the 36 scientists communicate and approach publics on social media? 37 We tried to answer this question conducting a first explorative research 38 related to the Digital Ethnography (Murthy, 2008; Coleman, 2010) in line with 39 the Rogers vision of Digital Methods (2009). We in fact extracted the material 40 directly connected with phenomenon observed making further a primary use of 41 secondary web data. Later the data collection we set a content quantitative 42 analysis of social media posts uploaded by selected subjects we observed on 43 two specific social media platforms during the first 3 months of the vaccine 44 dosing campaign in Italy, from December the 27th to March the 27th . 45 46

3 2021-4354-AJSS-SOC – 02 JUL 2021

1 Crisis Communication in Time of 2020 Pandemic 2 3 Through traditional and digital media, scientific communication has taken 4 on an important role during the Covid19 pandemic. In the specific case of Italy, 5 it increased the exposure of several scientific experts: virologists, 6 immunologists, and so on. who have provided to the large public important 7 scientific information about health, transmission, infection and, today, vaccines 8 program, but also well proper instructions to follow concerning the right 9 behaviour to adopt to face the pervasive contagion (Brondi and Pellegrini, 10 2021). The speeches, often discordant especially in the first phase of the 11 pandemic crisis, occurred on traditional media and social networks, with 12 different styles. 13 The Covid-19 crisis poses in fact significant challenges for how science is 14 conducted and communicated (Lasser et al., 2020). The 2020 agenda setting 15 «was substantially monopolized by the Covid-19 pandemic, the most 16 prominent feature in the news of the year» (de Sola Pueyo, 2021:1) driving 17 towards the infodemic dimension intended, following Hua and Shaw (2020) as 18 «the overabundance of information, sometimes not accurate, that creates 19 difficulties for society to understand which resources to use to access reliable 20 information» (de Sola Pueyo, 2021:1). 21 Following Wajahat Hussain (2020), in fact, through social and mass media 22 is possible to transmit a sense of unity due to large public coverage as well the 23 opposite: «Social media may also provide grounds for misinformation and 24 discrimination. People can utilize the flexibility and pervasiveness of social 25 media technologies to increase the public's adherence to the safety measures 26 suggested by global health organizations to combat the spread of covid-19». 27 Following Bucchi (1996) in certain situations, usually connected to 28 scientific controversies, scientists start to address the public directly by 29 skipping the usual stages of scientific communication in the way that «these 30 situations create a new modality in science communication that is associated 31 with different objectives and tactics compared to the traditional dissemination 32 pathways» (Olesk, 2021: 6). 33 According with this assumption, the best way to address directly the public 34 is through the architecture of digital scenario and the disintermediated 35 communication assets of the web environment as i.e. the social media sphere. 36 Social media platforms such as Twitter or Facebook ensure, in fact, 37 support and resilience between communities, «providing direct access to an 38 unprecedented amount of content and amplifying rumours and questionable 39 information» (Cinelli et al. 2020:1). 40 For Wajahat Hussain (2020), furthermore, the specialists are involved, in a 41 time of crisis as the pandemic, to turn their expertise communicating their 42 followers what’s happening and the situation overall in their premises at local, 43 National and International level. 44 Among these, of course, there must necessarily be scientists with strong 45 disclosure skills, called to deal with proper communication on social networks. 46

4 2021-4354-AJSS-SOC – 02 JUL 2021

1 Research Design 2 3 To better comprehend which scientists involve in our exploration among 4 the several ones who have progressively exposed disseminating their expertise 5 also through social media during 2020, we referred to a recent study of 6 Reputation Science, a research center specialized in crisis management 7 consulting particularly addicted to the scientific context. 8 9 Figure 1. Virologists’ Overall

10 11 Source: Reputation Science 12 This center synthesized a proper overall1 concerning the scientist visibility 13 on mass and social media from February 1st to November 20th, duly defined 14 rightly their public statements analyzed following two specific indexes: The 15 alert index that concerns the most frequent opinion of the scientist related to 16 the control and containment solution for the pandemic, while the coherence 17 index concerns the contradictions of their public statements along last year.

1Covid-19: The expert communication. Available at: https://www.reputationscience.it/analisi- dagli-esperti-italiani-sul-covid-19-sovraccarico-di-informazioni-e-indicazioni-incoerenti/

5 2021-4354-AJSS-SOC – 02 JUL 2021

1 Later to an in-depth inspection of the biographies of all the scientists 2 qualified in this overall we selected 5 of them rightly to their presence on the 3 social media. 4 5 The Selected Scientists 6 7 The first one selected is Roberto Burioni: an Italian scientist with an 8 international profile. He was, in 1988, Visting Scientist to Center for 9 Molecular Genetics to University of California of San Diego, while in 1991 he 10 was Visiting Investigator at the Department of Immunology of the Scripps 11 Research Institute La Jolla, California (USA) where worked in Dr. Dennis R. 12 Burton lab. 13 In 2004, Roberto Burioni worked at the Faculty of Medicine and Surgery 14 of the University Vita-Salute San Raffaele in while today he’s currently 15 Full Professor of Microbiology and , as well the head of an 16 immunological research laboratory. His researcher’ studies concerns the field 17 of development of human monoclonal antibodies against infectious agents. 18 Burioni became famous during the late 2010 years with his media 19 interventions, especially on social media on the issue of vaccines to counter 20 disinformation. His activity of scientific dissemination, always about vaccines, 21 switched also on TV allowing his reputation to gain more visibility. In 22 November 2018 he opened a website: Medicalfacts.it, dedicated to scientific 23 dissemination in the medical field. In 2019 he founded the Association "Pact 24 for Science" whose goal is to enhance the scientific evidence at the basis of the 25 legislative and government choices of all political parties. 26 He published several books of scientific divulgation and for this he wins 27 also several prizes. 28 From the beginning of the pandemic crisis still today, he appears as 29 regularly guest to the TV program “Che Tempo che fa” conducted by Fabio 30 Fazio which airs every Sunday at dinner time on the national broadcasting 31 service Rai3. 32 The second scientist involved in our exploration is Ilaria Capua, a 33 virologist of national and international fame. In 2000 she developed the 34 strategy Differentiating Vaccinated from Infected Animals (DIVA): the first 35 vaccination strategy against avian flu, by which - with a test - is able to detect 36 whether antibodies in a subject were induced by the vaccine or infection. 37 Ilaria Capua is responsible for an atypical action in the scientific field 38 dated2006: she challenged the system - obtaining international resonance - 39 deciding to make public the gene sequence of the avian virus. Then there was 40 talk of the birth of "open-source science”. 41 In 2013 Ilaria Capua decided to run for the Italian Parliament being elected 42 as the leader of her Civic Choice party. From May to July 2015 she was Vice- 43 president of the Chamber of Deputies in the Twelfth Commission (Social 44 Affairs). 45 She was put to criminal proceedings then acquitted (for conspiracy aimed 46 at the commission of crimes of corruption, abuse of office and illicit trafficking

6 2021-4354-AJSS-SOC – 02 JUL 2021

1 of viruses) which causes considerable inopportuneness in her personal life. In 2 September 2016, in fact, she decided to resign as a deputy and moved to 3 Florida where she got employed as researcher. 4 In the United States she heads a department of the Emerging Pathogens 5 Institute of the University of Florida. She later became director of the 6 University’s One Health Center of Excellence. 7 Furthermore, she is currently a resident guest at Dimartedi, a program of 8 the Italian Tv channel La7. 9 Furthermore, we involved also Fabrizio Pregliasco a researcher of the 10 Department of Biomedical Sciences for Health of the University of Milano 11 Statale. He’s author of expert reports for European registration of a vaccine 12 and flu medication. During his professional life he has collaborated in 12 13 clinical trials of vaccines and antiviral treatments. In 2015 he was selected as 14 Director of Health at the Galeazzi Orthopaedic Institute in Milan. Subsequently 15 he also became a consultant to the National Council of Economy and 16 Labour(CNEL), as well as to the National Council of the Third Sector (social, 17 economic and cultural reality in continuous evolution that includes bodies that 18 are neither public nor commercial). Both these roles are carried out bythe 19 Ministry of Labour. 20 From 2013 he’s president of ANPAS (an association committed to provide 21 public assistance). The efficiency in the activity of scientific dissemination is 22 the primary reason behind the esteem that the community of experts has 23 towards Fabrizio Pregliasco. During 2016 such an appreciation found concrete 24 form in the conferral of National Scientific Medical Union of Information prize. 25 During the emergency he is called to take on the role of 26 scientific supervisor. A role assigned to cope with the many deaths that 27 occurred in Milan at the Pio Albergo Trivulzio, an historic place of the city, 28 residence for the elderly, which accommodates over 1000 patients. 29 After the numerous deaths and at the same time the assessment of 30 responsibilities, his role helped Pio Albergo Trivulzio to implement a new 31 organizational structure. 32 He is sporadic guest in Tv programs on La7 and Rai channels. 33 Antonella Viola took place among the scientists observed in our research 34 also. She took a prize assigned in 2014 by the European Research Council a 35 scholarship: two and a half million of Euro in recognition of Steps project 36 considered a revolutionary program with regard to the evidence on immune 37 defenses against cancer. In the same year, she became associate professor in 38 General pathology at the Department of Biomedical Sciences of the University 39 of Padua. Today she is also a member of the scientific committee of the Italian 40 Association for Cancer Research, as well as being an auditor for the European 41 Commission dealt to the evaluation of scientific excellence projects. Thanks to 42 her contribution to molecular biology Antonella Viola becomes part of the 43 European Molecular Biology Organization.

7 2021-4354-AJSS-SOC – 02 JUL 2021

1 Finally, in parallel to her teaching and laboratory activities, she’s 2 responsible for promoting scientific dissemination, especially in the framework 3 of the European project Eufactor2. 4 The scientist is also particularly appreciated as a speaker: her clear style 5 leads her to travel the world as a speaker at conferences at prestigious 6 institutions. Between the most appreciated speeches are those at TED Talks. 7 She is a sporadic guest for TV programs on La7 Channel and on Radio 8 tune Radioradicale. 9 Finally, the last scientist selected is Alberto Zangrillo, head of the 10 Operative Unit of Anesthesia and General Reanimation and Cardio-Thorax- 11 Vascular, Head of Clinical Areas of the IRCCS San Raffaele Hospital in Milan 12 and collaborates at the La Madonnina Nursing Home. 13 He’s professor pro-rector and Full Professor of Anesthesiology and 14 Rianimation. Following Scopus3sources today he is one of the top ten doctors 15 in the world for the number of publications in the field of "anesthesia" and 16 "intensive care", author of since 800 publications, of which 400 on index 17 international journal that include randomized studies in The New England 18 Journal of Medicine, JAMA, Circulation and British Medical Journal. 19 His media experience is linked with the former Prime Minister Silvio 20 Berlusconi. Zangrillo has always been at his side, especially in the most health 21 difficult moments such as in 13th December 2009 when Silvio Berlusconi (at 22 that time Prime Minister) was hit by a small mermaid; or seven years later, 23 when the leader of the centre-right party was subjected to a decidedly complex 24 cardiac intervention at the San Raffaele. 25 By virtue of its authorial activity he collects numerous awards and honours 26 from the scientific community. Also the institutional offices seem to recognize 27 the merit and confer the merit titles by the Presidents of the Italian Republic, 28 Carlo Azeglio Ciampi and then Giorgio Napolitano. 29 He takes part sporadically to TV programs on Rai, Mediaset and LA7. 30 31 32

2The project of 2016 is aimed at young people between 16 and 19 years and was created to raise awareness of the study of science, technology and computer science, directing them towards training and professional paths that offer more opportunities, but they are often discarded because they are considered difficult or boring. The campaign also targets stakeholders and the general public, to draw attention to the importance of science and technology and to give visibility to the European Union’s commitment in these areas. 3SCOPUS is currently the largest bibliographic database of abstracts and citations of scientific literature. Index over 17,700 titles of scientific, medical, technical and humanistic journals, published by over 4,000 publishers. Among the most important citation functions Scopus allows to obtain: the H-Index or Hirsch Index (proposed in 2005 by Jorge E. Hirsch of the University of California at San Diego), is a bibliometric indicator that measures the impact of authors within the reference scientific community, based on number of publications and number of citations received; -to carry out the citation analysis of the authors and their relatives publications (through the Citation Tracker); to carry out the research and analysis of the authors' profile and membership affiliations.

8 2021-4354-AJSS-SOC – 02 JUL 2021

1 Methodology 2 3 Social platforms and data collection 4 5 As context unit, we selected two specific social media platforms: 6 Facebook and Twitter. 7 In a recent study, the research center Observa – Science in society (2019) 8 returned, in fact, interesting results relates about how Facebook turn out to be a 9 very suitable social platform by which Italian citizens are reached from 10 scientific-health centered contents and by which their fruition comes more 11 frequent, while on the other hand Twitter turn out to be the opposite. 12 13 Table1. Reading and sharing of contents concerning health and medicine (valid 14 % N=978 I read contents about health and medicine

Never Sometimes Frequently Total

Facebook 25,5 % 52,7 % 21,8 % 100 %

Twitter 67,4 % 24,6 % 8 % 100 % 15 Source – Observa science in society 16 17 According to this evidence, more scholars (Schultz, Utz and Goritz, 2011; 18 Eriksson and Olsson, 2016) argued about the perceived usefulness of Facebook 19 and Twitter in crisis communication. In particular, compared to other sources 20 related to the digital scenario, Twitter leads to less negative reactions than 21 blogs and newspaper articles, while Facebook results in a higher reputation and 22 less secondary crisis reactions than crisis communication via an online 23 newspaper (Eriksson and Olsson, 2016: 200). 24 On the base of this assumptions we started the data collection procedure on 25 Twitter availing of scraping procedure via Python syntax while for Facebook 26 we availed of the use of CrowdTangle, an insight tool reserved to the academic 27 hub that only tracks public available posts on Facebook, Instagram and Reddit. 28 The data have been collected following the structure of a proper standard 29 gather grid (Losito, 2003; Amaturo and Punziano 2013) divided in 4 main 30 domains (General information, Cross information, Engagement and 31 audiovisual and text information) and then organized in a Cases per Variable 32 Matrix composed by 1306 observations per 13 variables defined as follows: 33

9 2021-4354-AJSS-SOC – 02 JUL 2021

1 Table 2. Standard gather grid

General Cross Audiovisual and Text Engagement Information Information information

External Sources (No external Audiovisual Description source, Press, (No audiovisual elements, Account Institution and Data, infographics, Media (Viola, Burioni, Government, Like promotion, Normative alert, Capua, Science Journal, (Low, Medium and personal promotion, Press Zangrillo, Scientific High Likes) screenshots, Scientific Pregialsco) Network, publication extracts, Social Scientific Press, media screenshots, visual and Conference logo, Web events) Promotion, Official website) Repost Account Source (original post, Date Repost from (December, Press Institution Early January, and Comments Endo of Government, (No Comments, Low- Text January, Early Repost from medium comments and (Post corpus) February, End Scientific high comments) of February, network and Early March, Scientific End of March) Journals, Repost from other profiles) Tag (No Tag, Institution, Platform Press and other Share Text length (Twitter, profiles Tag, (Low, medium and (Short, medium and long Facebook) Scientific high share) length) network and Scientific journals Tag) Post Type (Tweet, Retweet, Tweet Photo and Video, FB Status, FB Status Repost,

Facebook Photo and video, Facebook Photo and video Repost, FB Link) 2

10 2021-4354-AJSS-SOC – 02 JUL 2021

1 The selected scientist proper profiles find place in the account variable 2 while date variable concerns the month when posts have been uploaded. All 3 contents have been classified due to the platform uploading (Twitter and 4 Facebook) and rightly of its classification unit (audiovisual or textual) duly 5 specified in the Post type variable by which furthermore come possible to 6 recognize the original or repost contents. 7 The external sources concern the context from which comes the external 8 links tied to the posts while the variable named Repost Account Source relates 9 the categorization of the account from where come the only reposts. All Tags 10 in posts have duly found place by a right categorization of the other mentioned 11 Social media accounts. For what concern the engagement, Like, comments and 12 share have been classified in tercile intervals that match low-medium and high 13 engagement levels, while pictures and video description have been categorized 14 by the symbolic representation of audiovisual material posted in audiovisual 15 description variable. Finally, we collected the text of every post as well the text 16 length properly classified following short-medium and long criteria cutting 17 tercile intervals. 18 All the information contained in the dataset have been processed following 19 a multi-stage analytical procedures consisting in the application of the topic 20 modelling aimed to point the features of the vaccination program discussion on 21 social media during the timespan we observed and then in the application of 22 the Multiple Correspondence Analysis (MCA) that make possible to detect the 23 latent dimensions by which mark the correspondence between topics and the 24 other context variables (platform, type of post, engagement and external 25 sources). Later, the Lexical Correspondence Analysis (LCA)4 turned out to be 26 the best way to a right synthesis of the collected data: by a compact graphic 27 representation of data relations projected on factors, we could point concepts 28 not previously observable by which find the right key-interpretations based on 29 the correspondence between the selected variables and most characteristics 30 words of post texts. 31 32 33 Analysis and discussion 34 35 The Topic modelling 36 37 The post texts present a considerable amount of information by which is 38 difficult to trace a semantic structure. For this reason, we offered the empirical 39 base to a simple but statistically robust solution: the topic modelling. 40 As first step we imported the database in T-Lab, a specific software 41 environment for the content analysis able to process proper patterns based on

4The LCA is a factorial technique concerning textual data and useful to: synthesize information contained in texts; make graphic displays of association networks among words and between words and texts; show the connections between text and context data. (L. Lebart, A. Salem, L. Berry, Correspondence Analysis of Lexical Tables, in Exploring Textual Data, in L. Lebart, A. Salem, L. Berry (eds.), Springer Sciences & Business Media, Dordrecht 1998, pp. 45-79.

11 2021-4354-AJSS-SOC – 02 JUL 2021

1 textual context. We submitted the text variable, consisting in the corpus 2 extracted by Facebook and Twitter, to T-Lab thematic analysis procedure that 3 is preceded first of all by the proper following automatic processes: 4 Lemmatization, consisting in 1) the standardization of all the verb forms in the 5 same mode 2) the transformation of nouns and adjectives posed in singular 6 number 3) the removing of definite-indefinite articles; Frequency threshold put 7 on 20 occurrences that drove us to exclude in the analysis all the words below 8 this frequency value and reducing finally the database to 650 total words; 9 finally we proceeded with the exclusion of empty segments not significant and 10 relevant for our analysis. Later we setup the modeling aimed to the topic 11 extraction based on the Latent Dirichlet allocation (LDA) a «generative 12 probabilistic model for text document collections based on a three-level 13 hierarchical Bayesian model, in which each item of a collection is modeled as a 14 finite mixture over an underlying set of topics. Each topic is, in turn, modeled 15 as an infinite mixture over an underlying set of topic probabilities. In the 16 context of text modeling, the topic probabilities provide an explicit 17 representation of a document» (Blei, Ng and Jordan, 2003). Following this 18 procedure, we extract 10 topics properly renamed respecting statistical criteria, 19 such as the consideration of specific words occurrences featuring the topic as 20 well the low-high shared words occurrences among all topics, and by the 21 semantic tagging (Bolasco, 2013: 126) on selected context in order to «detect 22 the right document meaning solving disambiguation and identifying concepts 23 by a set of words». 24 Finally, we classified the 10 topics taking account of its 985 emerging 25 elementary contexts intended as the document analyzed fragments in which the 26 topic itself comes more relevant. 27 Following Habert (2005), in fact, the more significance parts of documents 28 is supposed by the information weight of its fragments featured by its 29 discursive formulas, their position in the document, the specific weight of each 30 word related its scatter in the document etc. In our case, the resume of 31 elementary contexts T-lab returned us followed a hierarchical order based on 32 the informative score of single fragment. which text reduction has been 33 synthesized by 95 % threshold. 34 The emerged topics have been duly defined as shown in table 3: 35 36 Table 3. Topics Description Most frequent Words (Specific, Shared with Topic Name Elementary context examples high probability, Shared with low probability) Variante, Virus, Nostro, - Cos'è la variante sudafricana e perché Nuovo, Maggiore, ci preoccupa? Si tratta di una variante Pandemia, Mutazione, che presenta 3 mutazioni preoccupanti Virus Riuscire, Inglese, sulla proteina Spike la ormai nota Mutation Governo, Sudafricano, N501Y, comune anche alla variante Sars, COV, Ultimo, inglese, e le due nuove mutazioni Bambino, Portare, K417N e E484K Oltre ad aumentare la Rendere, Alimentare, trasmissibilità del virus.

12 2021-4354-AJSS-SOC – 02 JUL 2021

Difficile - a far saltare i numeri sono entrate in scena le nuove varianti virali La presenza delle varianti del virus nel nostro Paese desta giustamente preoccupazione, non solo per la loro maggiore trasmissibilità ma anche, nel caso delle varianti brasiliana e sudafricana - risposta a una singola dose di vaccini che devono essere somministrati in due Infezione, Dose, Paziente, dosi, ma purtroppo suggerisce che per Prima, Anticorpo, Usare, questi pazienti bisognerà pensare a Rispondere, Protezione, forme di protezione diverse Aspettiamo Approvare, Risposta, i risultati della seconda dose. Effectiveness Gravi, Domanda, Basare, - La protezione completa si stabilisce 7 of Vaccine Immunitario, Risultato, giorni dopo la seconda dose, quella Generare, Immunità, parziale 12 giorni dopo la prima dose Plasmare, Settimana, La notizia è "focolaio in RSA a Prato Sistema "Che fossero vaccinati una settimana prima o non vaccinati è completamente irrilevante - L'AIFA ha giustamente deciso di non consigliare l'utilizzo del vaccino AstraZeneca per gli over 55 Questo perché non ci sono dati solidi sull'efficacia del vaccino in questa fascia di età e perché è comunque un Vaccinare, Persone, vaccino meno efficace rispetto agli Astrazeneca, Popolazione, altri, riuscendo a proteggere solo 6 Decidere, Uscire, persone ogni 10 vaccinati La mia Somministrazione, AstraZeneca domanda però è la seguente che Ricevere, Problemi, Case facciamo con i 40enni diabetici? Rispetto, Evento, - Ricevo richieste da parte di chi è stato Contagio, Riguardare, vaccinato con il lotto AZ Accadere, Fiducia, temporaneamente bloccato da ISS Il Evitare, Lotto, Anziano lotto non è “ difettoso ” come mi scrivete la sospensione è precauzionale e se siete stati vaccinati non dovete fare nulla La febbre è comune con questo vaccino niente panico - - I vaccini funzionano contro le nuove varianti? Cominciano ad arrivare i primi dati Uno studio in revisione e i dati ottenuti da Moderna indicano che Dati, Efficacia, Rischio, gli anticorpi generati dai vaccini a Efficace, Clinico, mRNA continuano a riconoscere le Malattia, Studio, Sudi, varianti ma con minore efficienza Relevance of Moderno, Dimostrare, Moderna ha già annunciato di aver Data Bloccare, Sicuro, iniziato a generare una versione Soggetto, Ridurre, aggiornata del vaccino. Funzionare, Ottenere, - Ci eravamo basati sui dati pre clinici Comunità per dire che i vaccini a mRNA facevano ben sperare che anche l'infezione fosse bloccata e che saremmo rimasti in attesa di ulteriori

13 2021-4354-AJSS-SOC – 02 JUL 2021

dati che sono arrivati La vaccinazione di massa che è avvenuta in Israele e Regno Unito ci permette di rispondere a questa domanda basandoci sui grandi numeri e sui dati ottenuti in questi. Vaccinazione, Italia, - Io m'aspettavo un clima tutto diverso Paese, Noi, Sanitario, per questa campagna di vaccinazione Campagna, Scientifico, Preparation of Una mobilitazione febbrile, centri di Effetto, Scuola, Giorno, Vaccine vaccinazione aperti ventiquattr'ore, un Cittadini, Situazione, Campaign attivismo generoso Un'atmosfera Scelta, Trovare, Numero, potente, commovente, da trincea, da Alto, Grave, Prendere, ultima spiaggia Ricordare, Necessario - l San Raffaele non ci fermiamo neanche il giorno della Befana! Ecco il prof Ottavio Cremona, ordinario di Anni, Scienza, Mettere, Anatomia Umana Normale, che si Salute, Donna, Vita, sottopone alla vaccinazione anti Scientific Corriera, Ilaria, Capua, COVID 19 Università Vita Salute San Network San, Raffaele, Ospedale, Raffaele IRCCS Ospedale San Raffaele Gravidanza, Laboratorio, Gruppo San Donato. Marzo, Futuro, Umano - Orgoglioso di questo straordinario Team Multidisciplinare IRCCS Ospedale San Raffaele Università Vita Salute San Raffaele! - BioNTech, l'azienda che ha messo a punto uno dei due vaccini a mRNA, in settembre ha comprato da Novartis una fabbrica per dedicarla alla produzione del vaccino Ha impiegato 28 ( ventotto ) giorni per adattarla, 60 per cominciare a produrre concretamente il vaccino Nel dettaglio Settembre acquisto Tempo, Arrivare, Anno, azienda Ottobre novembre Produzione, Produrre, progettazione Inizio dicembre domanda Vaccine’s Possibile, Primo, Milioni, per adattamento. Supply Pensare, Caso, Ritardo, - Si parla con enorme interesse del Disponibile, Sicurezza, vaccino russo Sputnik, addirittura Differenza, Unico, Unito immaginando di produrlo noi e prima ancora di aver ottenuto l'approvazione da EMA o FDA Lasciando da parte interessi politici ed economici, non si comprende perché mai dovrebbe interessarci proprio questo vaccino e non quello di Johnson&Johnson Entrambi i vaccini si basano sullo stesso vettore adenovirale Pfizer, Notizia, Dosi, - ''Ottima notizia Dopo milioni di dosi Giorni, Proteggere, somministrate ( due anche a me ) Europa, EMA, Fase, l'EMA conferma la sicurezza del Response Sperare, Pubblicare, vaccino Pfizer BioNTech PS dopo la Capabilities Continuare, EU, seconda dose male al braccio, lieve mal Problema, Partire, Ottimo, di testa, dolori muscolari ma 15 giorni Morto, Italiano, Iniziare, fa dopo un'incauta partita a tennis stavo Insieme, Possibilità molto peggio.

14 2021-4354-AJSS-SOC – 02 JUL 2021

- Il vaccino Pfizer sembra proteggere dall'infezione già dopo una singola dose. Questa è davvero la migliore notizia degli ultimi tempi! Non sappiamo però quanto dura la risposta immunitaria quindi per il momento avanti con le seconde dosi nei tempi indicati e avanti coi vaccini! - Io avrei riportato le fonti, ma egualmente domani i giornali avrebbero titolato "rissa tra scienziati'', "i medici devono mettersi d'accordo tra loro prima di parlare "o il classico "qui Burioni, Roberto, Medico, nessuno ci capisce niente. Vedere, EricTopol, - n ogni caso già sono abbastanza Scientists’ Vaccine, Chiedere, conosciuto Ho cancellato il tweet e non reputation Leggere, Tweet, vado a letto con il timore i leggere Conoscere, Iene, domani gli articoli dei soliti giornali Correggere, Lavoro che mi infamano e gli sberleffi dei frustrati di Twitter che mi riversano addosso il loro odio, entrambe cose non importanti, ma non piacevoli Per cui ho cancellato il tweet. - chetempochefa Stasera ','tornerà a #CTCF da ','fabfazio Con il professore Covid, Parlare, Spiegare, parleremo delle ultime notizie sul Coronavirus, Testo, peggioramento dei casi covid in Italia e Professore, della situazione dei vaccini Vi Chetempochefa, Facts, aspettiamo dalle 20 00 su ','RaiTre. Media Medical, Emergenza, - Adnkronos #CovidItalia, ','''Varianti Presence News, Virologo, Gennaio, vera emergenza "Covid Italia, FabFazio, Pandemic, Pregliasco "Varianti vera emergenza Storia, Raccontare, "Secondo il virologo servono azioni Pericoloso mirate e tamponi nelle scuole "Le prossime tre o quattro settimane le più delicate' 1 2 Multiple Correspondence Analysis (MCA): The factors’ building 3 4 Further the topics, as active variables useful for the factor building were 5 pointed also engagement rates, platforms, type of post and date. 6 Scientists’ reputation and media presence topics characterized in the same 7 way both the two factors, while the topics Virus Mutation, Effectiveness of 8 Vaccine, AstraZeneca Case and Relevance of Data, instead, characterized the 9 first factor, posed on the horizontal axis, which reflects the argumentations 10 related to the vaccine clinical value assessment divided in two characteristic 11 parts. As shown by Fig. 1, on the right side, renamed “Discussion introduction” 12 in fact, we can notice low engagement and no topics emphasized on the factor, 13 while on the left side, renamed “Controversial discussion” we can notice high 14 engagement levels appeared on February turn and corresponding to 15 controversial and delicate topics as the withdrawal of AstraZeneca batches and 16 Vaccine effectiveness. For these argumentations Facebook and its relatively

15 2021-4354-AJSS-SOC – 02 JUL 2021

1 post types appears as the most used platform by the observed scientists, while 2 for the quieter discussion the elected platform is Twitter. 3 The second factor, posed on the vertical axis, is characterized by topics as 4 Vaccine campaign preparation, scientific network, vaccine supplies and 5 response capabilities. This dimension reflects the vaccination plan features and 6 is divided in two specific phases of our timespan of observation. First two 7 months in fact reflect the run up of the campaign while the last two months 8 refers to the follow-up and rating of the ongoing process of vaccine dosing 9 program. 10 The plan synthesizes how most of the observed subjects are more addicted 11 to Twitter posting practices compared to Facebook which is more preferred 12 only by Antonella Viola. The most of them prefer a social media exposition on 13 Twitter that, following Eriksson and Olsson (2016), connects the microlevel of 14 interpersonal communication, the meso-level of follower–followed networks 15 and the macrolevel of hashtag-based exchanges, while Facebook is usually 16 preferred for horizontal support among users during crisis situation as the 17 pandemic, in this case electing a posting-logic based on original contents 18 (Viola, Burioni and Zangrillo) or on shared post from other social media 19 accounts (Capua and Pregliasco). 20 Antonella Viola looks to be the only scientist open to controversial 21 discussion. Her position on plan, in fact, clears how she benefits of high 22 reactivity by her followers instead of Ilaria Capua and Fabrizio Pregliasco, 23 more oriented to quieter exposition about the vaccine argumentation, which 24 posts are in fact characterized by low likes, comments and shares. Lastly, 25 Roberto Burioni and Alberto Zangrillo communication is characterized by a 26 medium degree of reactivity of his followers. By the way, we must point that 27 the scientist of Istituto San Raffaele in Milan closed the comments options to 28 all his followers, even that for his following accounts or the ones he 29 spontaneously refers by a mention in the posts or in comments. 30 31

16 2021-4354-AJSS-SOC – 02 JUL 2021

Figure 2. Multi-correspondence Analysis

17 2021-4354-AJSS-SOC – 02 JUL 2021

The Lexical Correspondence Analysis (LCA)

Later to this first analysis procedure we more synthesized the information contained in our data providing the further explorative step of Lexical correspondence analysis (LCA) and show the multiple correspondences between words and context elements as for example who spread the message and the platform by which has been uploaded. Starting from the right side we can see reflected a relevant mass media exposition of the selected scientists. On both the up-down side are indeed marked words as: Dimartedi, the TV program that hosts Ilaria Capua as well as chetempochefa, the TV program that hosts Roberto Burioni every day at dinner time, duly complemented by the main character of their research network as , One Health UF and Elastico, the research centers and association tied to Ilaria Capua, as Eric Topol and Eran Segal, two co-author scientist of Roberto Burioni and Anpasnazionale, the association which Fabrizio Pregliasco is President. These contents, posted along the timespan we observed and crossing the discussion introduction about the vaccine campaign, reflect how the positioned scientist in this side follow a kind of hybrid disintermediation made by a social-mediatization process: they in fact disseminate via social media their statements originally conceived for mass media, finding on the web a new resonance chamber where spread the research outcomes to Twitter’s general public reluctant to react. On the other side, where the controversial discussion cross the timespan observed, a dialogic strategy is applied by only Antonella Viola. The high engagement levels in fact remarks an open possible discussion between the author and her public towards new scientific and mutable topics as the vaccines and its effectiveness due to the virus mutation as marked by words as: risk, worry, mutation, production, effectiveness, data, lockout etc. In this way, not relating with mass media frame, Antonella Viola looks as the only scientist truly according with a pure disintermediating process featured by contents thought specifically for the social media and the digital languages that promote a direct contact between sender and receiver, making outdated the figures deputed as intermediate in the communication processes.

18 2021-4354-AJSS-SOC – 02 JUL 2021

Figure 3. Lexical correspondace Analysis

19 2021-4354-AJSS-SOC – 02 JUL 2021

Main conclusions and further research perspectives

This first exploration has finally shown how the most accredited Italian scientists set their public exposition related to the new media sphere. The main hybrid approach, followed by most of the scientists observed, can be related to several factors on which focus later for further needed deepening that must be planned also by continuing to follow the vaccination program and its communication trends held abroad. A further comparison between Italy and other European Countries could be in fact useful in order to better comprehend these factors. More European Countries, such as France, Germany and UK, in fact, frame the communication of science availing of few experts officially related to Governments and Institutional research agencies. In this way, the Italian framing turns out to be more liberalized in terms of contribution for the public scientific debate, suggesting that the distance from a pure disintermediated digital vocation could not only be related to the idea of Italy as a laggard Country in terms of adaptation to the innovation in communication, while as sociopolitical context where arises a particular relation between science and politics. In this way, for example, we can wonder if the political sphere influences the scientific communication, scientists’ interests, strategies and exposition styles related to this fundamental step fighting the pandemic. Considering what has been shown, a further research perspective can be for example oriented on the base of the following hypothesis examples: The social media use of Roberto Burioni seems to be as resonance chamber of the framing processes he leads in TV, while Ilaria Capua adapt her twitter account as a promotion space while her Facebook account as a press office managed by third people. Do they don’t address controversial argumentation to protect their reputation and avoid troubles with the public broadcasting service? Does Pregliasco do the same to stay far by any ideological or political polarization misunderstanding rightly of his government collaboration? Reading the Science Reputation overall, Zangrillo has been qualified as the most contradictory scientist. Does he keep a low profile to avoid any other public embarrassment? In conclusion, in light of this first exploration, becomes needed a more required deepening of the scientific communication frame worked during the pandemic era, maybe starting from the above mentioned example-questions. Moreover, since we’re talking about a current and an in progress mutable phenomenon, we’re forced to press further on our investigation following and observing the related events that feature the vaccination issue due to the multiple surprises and releases that the vaccinal program can reserve in terms of scientists exposition in the public debate.

20 2021-4354-AJSS-SOC – 02 JUL 2021

References

Amaturo, E. Punziano, G. 2013. Content analysis: tra comunicazione e politica. Milano: Ledizioni. Ancarani, V. 1996. La scienza decostruita. Teorie sociologiche della conoscenza scientifica. Milano: Franco Angeli. Blei. D, Ng, A. Jordan, M. 2003. Latent Dirichlet Allocation. Journal of Machine Learning Research 3, 993-1022. Bolasco, S. 2013. L’analisi automatica dei testi. Roma: Carocci. Brondi, S. Pellegrini, G. 2021. Comunicazione pubblica della scienza: fonti, fiducia e credibilità. I risultati di un’esperienza di coinvolgimento dei cittadini, in Saracino, B. Pellegrini, G. Annuario Scienza Tecnologia e Società, : Il Mulino. Bucchi, M. 1996 La scienza e i mass media: la fusione fredda dei quotidiani italiani, in Nuncius, 2, pp. 581 – 611 https://doi.org/10.1163/182539196X00060.. Bucchi, M. 2006. Scegliere il mondo che vogliamo. Cittadini, politica, tecnoscienza. Bologna: Il Mulino. Bucchi, M. 2010. Scienza e società. Introduzione alla sociologia della scienza. Milano: Raffaello Cortina Editore. Bucchi, M. Trench, B. 2014. Handbook of Public Communication of Science and Technology. London: Routledge. Bucchi, M. Saracino, B. 2021. Annuario Scienza, Tecnologia e Società. Bologna: Il Mulino. Cinelli, M., Quattrociocchi, W., Galeazzi, A., Valensise, C.M., Brugnoli, E., Schmitd, A.L., Zola, P. and Scala, A. 2020. The COVID-19 social media infodemic. Scientific Reports, 10, 16598, https://doi.org/10.1038/s41598-020-73510-5. Coleman, E.G. (2010). Ethnographic Approaches to Digital Media, Annual Review of Anthropology 39(1), 487-505. de Sola Pueyo, J. 2021. Science in the media: the scientific community’s perception of the COVID-19 media coverage in Spain. JCOM 20 (02), A08. https://doi.org/10.22323/2.20020208. Eriksson, M. Olsson, E. 2016. Facebook and Twitter in Crisis Communication: A Comparative Study of Crisis Communication Professionals and Citizens. Journal of Contingencies and Crisis Management, V24, Issue4, doi.org/10.1111/1468- 5973.12116. Jasanoff, S. 1997. Science at the Bar. Law, Science, and Technology in America. Cambridge: Harvard University Press. Habert, B. (2005). Instruments et ressources èlectroniques pour le francais, Orphys, Paris. Hua, J., and Shaw,R. (2020). Corona Virus (COVID-19) “Infodemic” and Emerging Issues through a Data Lens: The Case of China. International Journal of Environment Research and Public Health, 17(7), 2309; https://doi.org/10.3390/I jerph17072309. Hussain, W. 2020. Role of social Media in COVID-19 Pandemic. The International Journal of Frontier Sciences, 4(2), 59-60, doi.org/10.37978/tijfs.v4i2.144. Lasser, J., Ahne, V., Heiler, G., Klimek, P., Metzler, H., Reisch, T., Sprenger, M., Thurner, S. and Sorger, J. 2020. Complexity, transparency and time pressure: practical insights into science communication in times of crisis. JCOM 19 (05), N01. https://doi.org/10.22323/2.19050801. Losito, G. 2003. L’analisi del contenuto nella ricerca sociale. Milano: Franco Angeli.

21 2021-4354-AJSS-SOC – 02 JUL 2021

Mazzoleni, G. 2004. La comunicazione politica. Bologna: Il Mulino. Murthy, D. (2008). Digital ethnography: An examination of the use of new technologies for social research. Sociology, 42(5), 837-855. Olesk, A. 2021. The types of visible scientists. Journal of Science Communication 20(02):A06, - DOI: 10.22323/2.20020206. Patel, K. 2013. Incremental journey for World Wide Web: Introduced with Web 1.0 to recent Web 5.0 – A survey paper. International Journal of Advanced Research in Computer Science and Software Engineering, 3(10), 410–417. Pellegrini, G. 2018. Narrazioni di mondi possibili. Giovani e immaginario scientifico. Bologna: Il Mulino. Pellegrini, G. Saracino, B. 2019 Annuario Scienza, Tecnologia e Società. Bologna: Il Mulino. Peters, P.H. Allgaier, J., Dunwoody, S. Lo, Y. 2014. Public communication of science 2.0: Is the communication of science via the "new media" online a genuine transformation or old wine in new bottles? In EMBO Reports - July 2014 DOI: 10.15252/embr.201438979. Rogers, R. 2009. The end of the Virtual: Digital Methods. Amsterdam: Amsterdam University press. Rogers, R. 2013. Digital methods. Cambridge: MIT press. Schultz, F. Utz, S. Goritz, A. 2011. Is the Medium the Message? Perceptions of and Reactions to Crisis Communication via Twitter, Blogs and Traditional Media, Public Relations Review 37(1):20-2, 2011 - DOI:10.1016/j.pubrev.2010.12.001. Scamuzzi, S. Tipaldo, G. 2015. Apriti scienza. Il presente e il futuro della comunicazione della scienza in Italia tra vincoli e nuove sfide. Bologna: Il Mulino.

22