Keywords

Digital Sociology

The changes that we have been witnessing for several decades are remarkable, starting with the birth of the Internet in 1969, going through the formation of the World Wide Web in the 1990s, to the mid-twenty-first century, which saw the birth of the first social networks, up to the consolidation of the so-called Web 2.0.

These developments have affected a wide range of scientific fields, and social sciences have also been called upon to face major challenges from an epistemological, theoretical and methodological standpoint (Savage & Burrows, 2007; Ciotti & Roncaglia, 2008; Ruppert et al., 2013; Mayer-Schönberger & Cukier, 2013; Kitchin, 2014a, 2014b; Marres & Gerlitz, 2016; Agnoli, 2016; Conte, 2016; Salganik, 2018). This led to the first articles and texts dealing with issues and topics specific to digital sociology (Wynn, 2009; Neal, 2010) until its academic institutionalisation in 2013 with the curatorship of Orton-Johnson and Prior entitled Digital Sociology: Critical Perspectives (2013). The expression “digital sociology” has now become part of the cultural lexicon as well as being specific to the discipline, which, as a branch of culture, communication and media sociology, is attracting a growing interest in the international context (Marres, 2012, 2017; Lupton, 2014).

Digital sociology is oriented towards considering two complementary aspects: examining the digital sphere not only as an object or technology but also as a field of study, considering what is happening as a real paradigm shift (digital turn) that the social sciences must take on as part of their purview (Caliandro & Gandini, 2019, pp. 17–18). This is an expression that has become increasingly used to indicate the new directions of sociological research towards digital technologies, which has focused its attention on the implications of technology on what happens in our lives, with interest in the effects on, among other things, the relational dynamics of social life, behavioural models, the creation and definition of the self, the spread of forms of online social interaction and how social relations, consumption practices and inequalities change; the contexts in which they are used are also increasingly wide-ranging, e.g. consumption, work, local transport (Uber), hospitality (Airbnb) and industry.

It is not our intention here to reconstruct the rich debate that has developed on these issues but to highlight just a few elements that are useful for the discussion. In this context, it is a shared opinion that since then, people have really started to talk about big data, digital research and web social science, and, from then on, many have become interested in and expressed their views on this epochal change. The changes and challenges that scientific research has undergone and continues to undergo today have highlighted how, on the one hand, the empirical opportunities have increased and, on the other, the risks have increased as well. The reactions—as Veltri (2021, pp. 1 et seq.) summarises—can be identified with two prevailing positions: scepticism and enthusiasm. More specifically, the first position dealt with arguments about data quality, access and ownership, as well as the integration of online and offline data; the second position focused mainly on the fact that the wide availability of digital traces represented a breakthrough in the increasingly difficult collection of data. In addition to these two positions, there was also an intermediate proposal, somewhere between the two. A shared fact is that digital data has changed the way of doing science and research, so much so that even researchers who welcomed the shift from analogue to digital data with an initial enthusiasm related to the great potential were later involved in a reflection on research methodology.

The approach to the digital domain as a field of study arises primarily as a methodological question, in relation to the large amount of data available. An important issue here is the value of data. Its increasing obsolescence, caused by the rapidity of the whole research process, has meant that it has become cheaper, and, consequently, there has been a decrease in the quantity of data required for the planning and preparation phases of the analytical strategy.

The issue soon became also theoretical—as mentioned above—in the sense of an overall approach to the study of the digital sphere. Today, in fact, social action is closely interconnected and integrated with digital technology, so much so that digital contexts have become real social places from which society can be observed and measured (Marres & Gerlitz, 2016). A multitude of the social relationships in which we are involved take place in or are mediated by digital contexts. In this way, digital sociology has developed itself not only on methodological and theoretical levels but also on epistemological ones. From this last standpoint—as suggested by Caliandro and Gandini (2019, p. 20)—there are three relevant issues: the need to add to the media-centric perspective of cultural and media sociology a perspective that puts the social actor at the centre (see Jurgenson, 2011), the rejection of the conceptual and epistemological opposition between online and offline and the rejection of a technical-deterministic conception of digital media. From this perspective, studying the digital society means conceiving digital media and the social relations that take place in them in a unified framework, in a mutually inclusive vision, researching technology and society in a unified vision.

The perspective proposed by digital sociology stands as an alternative to the traditional dichotomy of qualitative and quantitative approaches (Natale & Airoldi, 2017), but not in terms of opposition but rather of complementarity, since it is “capable of supporting, triangulating and completing the study of social action in the Internet age” (Caliandro & Gandini, 2019, p. 15). The Internet is an object of analysis and a source of methods (Rogers, 2009).

Digital Methods

It is within the context of the debate outlined above that, starting from the well-known work of Richard Rogers Digital Methods (2009), digital methods (which traditionally straddle different disciplines, including media studies, digital humanities and social sciences) assert themselves as approaches, tools and methods suitable for studying digital environments with the aim, not so much of virtualising traditional research methods by adapting them to the virtual world and understanding how much culture and society are present online but, by overcoming the real-virtual dichotomy, of going beyond the study of online culture to study sociocultural changes using the Internet. In this regard, it is Rogers himself who indicates the importance of immersing oneself in the digital world under observation, “following the medium”, following the technical strategies that the same environment uses in order to collect, organise, examine and structure the communication data flows of which it is made up. In this regard, digital methods should be considered as approaches, tools and methods suitable for studying digital environments in a “native” mode, which refers precisely to the natural way in which digital data is organised and which the researcher can use to study the social relations of the digital environment, the ways in which they are formed, maintained or ended, as well as the social and semantic structures that denote them (Caliandro & Gandini, 2019, p. 12). The researcher who moves in the digital world and with digital methods must bring to light networks of meaning (Geertz, 1973) and sociality that social actors and digital devices weave, for example, within social media around a specific object (brand, political issue, celebrities, etc.). Therefore, he or she must be oriented not so much towards identifying an online community but rather mapping the practices through which users and devices construct social formations around a moving object (Büscher & Urry, 2009).

The debate on the changes that characterise social research with the widespread use of the Internet, in general, and with the availability of so-called big data, in particular, has been open for many years now. Big data has gradually been defined in different ways and among the characteristics associated with it there are certainly the well-known “3vs” (Laney, 2001; Zikopoulos et al., 2011; Lombi, 2015): volume, which corresponds to the amount of data generated; variety, which concerns the diversity of formats and also refers to their possible lack of structure (different documents, social networks, microblogging platforms, etc.) and different sources (automatically generated, user-generated, etc.); and velocity, the speed with which data is made available. Big data can also be distinguished by further characteristics: exhaustiveness, which has to do with the difficulties of selecting a sample and its true representativeness; high resolution, resulting from the combined use of more defined visual data and the processes of indexing and tracking people and objects; relationality, the opportunity to link and aggregate data in order to create an amplifying effect (see Crampton et al., 2013); and flexibility, which refers to its adaptability and is considered by many to be the most interesting characteristic from a social science perspective (Aragona, 2016, pp. 44–45).

In the light of this, it can be argued that the characteristics of digital data have drawn attention to the need to use analytical approaches other than those defined as conventional in social sciences. Among quantitative analyses of digital data, three types can be identified: statistical methods of reducing the dimensionality of data, quantitative methods of analysing relational data (network analysis) and quantitative methods of analysing texts as unstructured data (including sentiment analysis). Finally, it should be pointed out that the focus on these categories of quantitative analysis does not mean that conventional descriptive and inferential techniques are no longer used; what is being pointed out is the fact that the nature of digital data makes these three techniques more common (Veltri, 2021).

With what has been said so far, the foundations have been laid for the development of the dialogue that lies at the heart of this intervention: the dialogue between modernity and tradition. Starting from the computational approach mentioned earlier, we have seen that three types of analysis have been identified. Two of these (sentiment analysis and network analysis) are those that will be compared with a proposal from the traditional theoretical landscape of social sciences (the themes of “personal influence” and the “two-step communication flow” theory).

Social Network Analysis

The social network analysis (SNA) originated with Jacob Levi Moreno’s graph theory (1953). It stands as a variant of structural sociology that considers networks (relationships made up of contacts and exchanges), looking at the non-episodic relationships between social persons they delimit or create action opportunities for individuals.

The measurement of networks gives rise to data that is different from that of other social sciences and therefore a set of methods for its analysis has emerged (Barabási & Posfai, 2016), which has found wide application in the field of digital research; closely linked to this is the issue of reconstructing the networks generated by the use of social media. To this end, the footprints that each individual leaves behind are of the utmost importance, since it is possible to reconstruct links and identify strategic nodes from these; in fact, with the disclosure of one’s opinions on the web, a mass of data (big data) has been created that is useful from various points of view (of institutions, political figures, brands, etc.) because of the impact it can have. In this regard, the usefulness and applications of SNA are many: reputation; perception of brands, characters and products; measurement of social media marketing activities; degree of satisfaction; improvement of services and products; prevention and management of online crises; identification of competitors; and identification of influencers.

It is clear that the research methods used in digital social research are not exclusive to SNA, but the same plays a very important role. Similarly, the analysis of online networks does not coincide with the study of social media, since not only the data that can be deduced from the latter, but all that corresponding to our digital traces (telephone calls, commercial transactions, movements signalled by GPS, etc.) can be analysed in the form of networks, and they represent an epistemological approach that is well suited to understanding a hyperconnected world (Easely & Kleinberg, 2010; Riva, 2011; Kramer et al., 2014). As Veltri (2021, pp. 125–127) suggests, at the basis of network analysis, there is an important change of perspective compared to what is usually done in social and behavioural sciences, where we are used to dealing with data by arranging them in matrices of cases by variables, assuming the independence of observations on individual units; therefore, data for different subjects does not depend on each other. This orientation, which is typical of methodological individualism, contrasts with that adopted by network analysis, namely, the constructivist-relationalist approach; in fact, SNA is interested precisely in the interrelationship between individual units, i.e. the relationships between subjects. Network-based theories conceive units as elements that do not act independently but influence each other; to test these theories, the measurement of networks is based on the use of structural or relational information.

Web Sentiment Analysis

In this context, one of the analysis tools that have become more widespread with the massive extension of social media is the “sentiment analysis” (Pang & Lee, 2008) (also called “social media analysis” or “web sentiment analysis”) (SA), which is now being studied in both academic and commercial contexts. In this respect, it should be pointed out that SA is part of the more general content analysis (text mining) and that there are both qualitative and quantitative forms of content analysis. Specifically, it concerns the platforms, the expression of users’ judgements and the computational analysis of feelings and opinions expressed within texts generated on the network by the interactions between users, concerning a product, a service, an individual, an event, etc., in a given space and time frame.

If we think about the era in which we live, the era of digital data, we all know that the texts generated by each user are growing by leaps and bounds and this presents a challenge in terms of the ability to analyse these quantities, which requires a combination of computational and conventional textual analysis methods. SA has found a certain application in the social sciences, becoming an interesting field of study, and its diffusion is closely linked to the availability of digital data in relation to opinions, assessments and judgements; one need only think of social media platforms, blogs and everything else that represents a vast source of data relating to people continuously expressing opinions on people or objects of the most varied kinds. On the one hand, it is a very widespread tool, just think of the numerous applications for its implementation made available by social media themselves (Facebook Insights, TweetStats, Google Analytics, etc.), as well as the availability of paid platforms offering this type of service/analysis (Talkwalker, Digimind, etc.). On the other hand, the methodological soundness of this type of analysis is frequently questioned, especially when compared to traditional (textual and other) analysis tools and techniques, such as telephone surveys and focus groups. Criticism and diffidence have mainly focused on the context, emotional ambiguity, sarcasm and the polysemous nature of language, and although the algorithms underlying the tests have not been able to predict with very high accuracy the feelings associated with people (Saif et al., 2013, 2014; Bravo-Marquez et al., 2014; Vora & Chacko, 2017), this is a field that is finding wide application and growth in many areas.

This issue has recently been addressed in the literature of the sector, which, besides having dealt with some methodological issues concerning the use of digital footprints and the data provided by them (accessibility in view of commercial and/or legal limitations, ethics in their free use, representativeness), has highlighted how the two perspectives (traditional and modern), instead of being opposed, can be used in an integrated manner.

Tradition and Modernity. Comparing Perspectives

“Personal Influence”

In this framework, a further element of interest, which is the focus of this contribution, concerns the comparison between the scenario offered by the analysis of digital footprints through the above-mentioned network and sentiment analysis and the well-known “two-step communication flow theory”. Expounded for the first time in 1944 by Paul Lazarsfeld and subsequently revised with Elihu Katz in 1955, this theory addresses the methodological, as well as substantive, problem of studying the effects of mass media, investigating both the effect of communication between people and on people. In addition to proposing a revision of the approach to mass media that takes into account the individual, the web of ties that he or she weaves, the small groups in which he or she participates and the social conditioning on the existential dimension of the subject, the study goes further, offering fundamental contributions regarding the pervasiveness of conditioning, the network structures of such conditioning, highlighting the subtle tools that it uses and the strength of the drive that it has towards the internalisation of values (Ferrarotti, 1968, pp. VII et seq.).

The expansion of mass media (TV, cinema, radio, press) described by Katz and Lazarsfeld refers to technical innovations affecting the specific sphere of culture; it is analogous to the industrialisation process affecting the economic and labour spheres. Both contexts are thus involved in a process of rationalisation. However, while the phenomenon of industrialisation is structurally oriented and therefore outlines linear research vectors, on the other hand, the role played by mass media is placed at a superstructural level and, by virtue of its greater complexity (influences intersect with action, full of stimuli and proposals), leads to more complex methodology issues (Ferrarotti, 1968, p. IX).

The aim of the research is to understand the effects of mass media, measuring their influence on people’s behaviour and psychology. Specifically, there were two directions followed in the research of the time: the first was typical of public opinion surveys and was characterised by the search for correlations between social factors and changes in circulation; the second was the search for “pure” effectiveness, the ideal flow from the media to the public, directing its opinions. It is clear that the critical debate on this last approach has been and still is full of doubts and stimuli, especially in the direction of the actual feasibility of an intent such as that just described.

One of the central points, in terms of its originality, of the research proposed by the authors of Personal Influence is the role accorded to the small group, placed between “communication” and “mass”, with the web of interpersonal relations that constitute it. In this sense, both from the point of view of theory and research, attention is drawn to the connection between the study of the informal group and the study of the effects of mass communication (Statera, 1968, p. XXX). The orientations within the study of the effects of mass media in those years were characterised by two opposing positions: those who saw them as a democratic instrument giving everyone equal access to information and those who saw them as diabolical agents aiming at the destruction of democracy. Far from being in contrast—as Katz and Lazarsfeld point out in the beginning of their book—the two conceptions of the function of mass media actually have the same image of the mass communication process: “a new kind of unifying force” that extends and reaches out to every individual “in a society characterised by a scarcity of interpersonal relationships and an amorphous social organisation” (1955, p. 4).

Precisely interpersonal relations have been accorded such centrality that someone has spoken of the “discovery of ‘people’“ (Katz & Lazarsfeld, 1955, pp. 11 et seq.), attributing to interpersonal relations the role of a variable intervening in the mass communication process. Research shows that opinion formation and possible change are linked to a specific factor: personal influence. Individuals are subject to the influence of the people they come into contact with in small groups (family, colleagues, friends, etc.), among whom some are characterised as opinion leaders; the latter, in turn, are influenced by mass media to a greater extent than non-leaders. From this dual process comes the idea of the two phases of the communication flow. “In short: ideas seem to pass from the radio and from the press to the opinion leaders and from them to the less active sectors of the population” (Katz & Lazarsfeld, 1955, p. 16).

In this framework, we outline how opinion leaders are present in all social and economic strata and how they are an integral part of the interaction that takes place on a daily basis in personal relations; therefore, interpersonal relations are potential communication networks within which the opinion leader, as a member of a group, plays a fundamental role of communication towards the other members with whom he or she is in contact.

The research project carried out by the two authors is a complex one, which is why we have chosen to deal here only with those topics that are relevant to what we intend to discuss. The definition of communication flow networks has highlighted some important concepts that—as will be shown later—are well connected to some tools used in the field of digital methods and mentioned above. To this end, and in addition to what has already been said, it is worth drawing attention to the fact that, in the work plan drawn up by Katz and Lazarsfeld, space was devoted to an in-depth examination of the structural connections inherent in the social ties between individuals and how they influence interpersonal communication. In this intention to delve into interpersonal transmission models, the reference to Moreno’s sociometric technique (1953) is made explicit directly by the authors. A subsequent step is aimed at identifying strategic transmission roles in the interaction with others and at this point of the work Leavitt’s concept of “centrality” (1952) is recalled to describe the degree of access of a subject to all the other subjects in the group, linking it to the role of leader.

Moreover, it should be pointed out that the two American authors were specifically interested in the analysis of political propaganda and commercial advertising; digital methods are also said to have originated as a qualitative-quantitative approach mainly functional to the study of political issues emerging on the web (Rogers, 2013). However, beyond the investigated phenomenon, it has been seen that the research highlighted a structure of the influencing process, which we defined above as the ideal flow (effectiveness) going from the media to the public and directing their opinions. Over the years, this proposal has had a rich series of applications and revisitations (Postmes, 1997; Liu, 2007; Park, 2018), among which we specifically mention Jensen’s (2009).

As can be seen from the figure below, the proposal draws a parallel between the two-step model of 1944 and its 2009 version. The latter is clearly a modern adaptation to the specific context of digital social networks, which are really the third step added to the original version. What changes further are the individual contacts with the opinion leaders: while in the original model these are undirected contacts, in the revised version, they are not only directional but even two-way (Fig. 19.1).

Fig. 19.1
2 flow diagrams. Diagram 1 has 2 steps. In step 1, the node for mass media is divided into 3 with 2 small circles, each of them is further divided into 3 smaller circles in step 2. Diagram 2 has 3 steps. In step 1, the node for mass media is divided into 3 small circles. Step 2, 3 nodes for digital social networks, each with 3 further divisions in step 3.

Two-step flow (Lazarsfled et al., 1944) and three-step flow (Jensen, 2009) comparison

Full size image

Katz and Lazarsfeld proposed a two-step communication pattern that was witnessed in different contexts (industrial, urban community, etc.). That model, in a version adapted to digital communication, is still present and valid today, with a few additions. In short, we could say that the medium has changed and the modes have adapted: in the past, the conditioning of opinions took place through traditional media, such as TV; today the stage has changed to social media. In this respect, it is fair to note that the changes in communication modes offer a much more interesting panorama in some respects.

Influencer Logic

Although the expression “digital society” refers to a vast complex of relations, objects, environments and actors and although social media are not an exclusive part of the analysis of the ways in which social actors act in forms mediated by technology, it has been said that social media, as social spaces, represent one of the primary objects of interest of digital sociology (Caliandro & Gandini, 2019, p. 20) and within them the activity of figures identified as “influencers” takes place. Whereas in the past, speakers, communicators, political figures and leaders used the streets and then the traditional means of communication—those studied by Katz and Lazarsfeld, for example—to speak to the masses, today’s influencers use social media.

At this point in the work, we could say that “the concept of opinion leader and that of influencer are not far apart”, aware that this would provoke a not very positive reaction, not to say disconcertment. Well-aware of the importance of the Katz-Lazarsfeldian tradition and of the undeniable disruption with which modernity pervades our lives, as individuals and as researchers, we cannot deny—even though some scholars consider it lacking in scientific depth—the notion, the figure and the role that influencers play nowadays. As a first step, we would like to draw attention to the fact that when we talk about influencers, a variety of people who have become known through social media come to mind, and very often we think of people who are the object of the most disparate criticism, on the one hand, and of the most total veneration, on the other. In this respect, however, it is worth emphasising that this ambivalent figure has a central role and exerts its presence with respect to a plurality of interests (such as the political sphere, the measurement of reputation, the assessment of marketing activities) and in a variety of contexts (tourism, fashion, sports, food, etc.) (Senft, 2008; Ejarque, 2015; Viviani, 2017; Del Franco García & Segado Sánchez-Cabezudo, 2016; Olietti & Musso, 2018; Braves et al., 2019; We are social, 2022).

In the virtual world, one can most often read and listen to opinions on this subject, and many times—whether rightly or not is not for us to determine here—other issues are called into question (sexism, minorities, power, etc.). The debate is mostly characterised by regarding the word “influencer” as a synonym for stupid, superficial; to call someone an influencer is to mock them and belittle what they do. Someone else countered by saying that this happens “since this is an activity in which women are on average more successful”. Others point out that, in reality, influencers require more skills than one imagines: a good knowledge of marketing, video-photographic skills, writing and storytelling abilities, as well as qualities such as empathy and the ability to create a bond with the audience, continuous updating and constancy.

First of all, however, it should be borne in mind that influence is a complex phenomenon closely linked to cultural, psychological, economic, communicative, political and social aspects. In the age of digital media, studying and understanding influence and how it works has become increasingly complex. In such a framework, the issue of digital influence, taking into account the various points of observation, has been the subject of in-depth studies that have led to numerous studies and theorisations on the subject, at the centre of which is the figure of the influencer (Polesana & Vagni, 2021), which, therefore, leads to the creation of a true “influencer logic”.

But what is the meaning of the word “influencer”? Actually, influencers are those who have a large following on social media, regardless of the topics they talk about (make-up, sustainability, science), and, unlike any other user, influencers are able to influence their numerous followers with their opinions, generating strong opinion trends, so much so that the opinion of an influencer cascades over a number of other online users who largely align themselves with their point of view. For example, influencer marketing uses this effect to give maximum visibility to a brand or product, accompanying it with a halo of acceptance. Having an influencer who speaks well of a product means gaining the interest and attention of at least one segment of followers, who will be inclined to perceive and/or evaluate the brand or product in a positive way precisely because the influencer has pointed it out and spoken positively about it.

Are there any characteristics that make it possible to identify an influencer? The main ones are five:

  1. (a)

    High number of followers. There are different classifications of influencers based on the number of fans, one of which, by way of example, which is considered valid in the Italian context, classifies them into micro influencers (from 5000 to 20,000 followers), middle influencers (from 20,000 to 150,000 followers) and top influencers (above 150,000 followers). Niche areas, such as those with a strong technical and scientific characterisation, should be excluded from these groupings, for which the numbers are certainly lower.

  2. (b)

    Original and relevant content. Influencers are characterised by publishing valuable content with a certain ability to address topics and an approach strongly characterised by a personal style, clarity, incisiveness and empathy.

  3. (c)

    High engagement. One of the main characteristics that distinguishes an influencer is the ability to dialogue and interact with the audience, so as to be a reference point for the latter in creating opinions. This can be detected through some typical social traits, such as instant interactions (likes, tweets), shares, comments and replies to comments.

  4. (d)

    Charisma and identifiability. The influencer is distinguished by being a person with a strong character who makes the way something is said almost more important than the content itself.

  5. (e)

    Ability to persuade. A characteristic trait is to be able to induce followers to follow one’s suggestions. An influencer differs from a simple columnist in that he or she does not just give advice or propose ideas but is able to influence the choices and actions of followers, inducing a positive reaction in terms of adherence (comments, reactions).

The above characteristics should be understood in combination with each other. Thus, for example, any classification of influencers based solely on the number of followers is not effective; it is necessary to consider the type of relationship, the level of consideration that followers have for the influencer. Furthermore, with regard to engagement, a post addressed to a pool of tens of thousands of users may receive no reaction, which denotes little influence on followers, while a post that receives many comments but no counter-response from the influencer indicates a lack of relational availability, which in turn implies a lower capacity to influence. For this reason, many companies relying on influencer marketing increasingly prefer to focus on micro influencers, who have a smaller number of followers but are still able to interact directly with them. This is the reason why companies often prefer these influencers, because they are more likely to create a network of direct relationships. Therefore, they are characterised by greater credibility and, consequently, their ability to influence users’ decisions is greatest. On the other hand, influencers with a large number of followers most likely will not be able to relate directly to their fans, thus losing their persuasive effect.

Concluding Remarks: Personal Influence vs Influencer Logic

In the digital age, there are an increasing number of areas where the footprints we leave behind (voluntarily or not) become relevant for the use (legitimate or not) that can be made of them, creating new broad scenarios of analysis in different fields of interest: politics, stock markets, communication and marketing, sports and medical and natural sciences. By surfing the web from any device (mobile phone, tablet, personal computer), each of us now leaves footprints. As explained above, this changing landscape has posed and continues to pose challenges for social sciences, which are called upon to reflect on new issues from both a theoretical and empirical perspective. In this regard, the use of research tools, such as those discussed in this paper, poses many questions to the researcher regarding their robustness, also in comparison to traditional research methods and techniques. The diagram in the figure represents the central focus of this exposition (Fig. 19.2).

Fig. 19.2
A diagram compares 2 step flow communication and digital methods. 1, personal influence, pre-existence of work, small groups, survey content analysis, offline, few communicators, and many listeners. 2, influencer logic, co-construction of the network, nodes hubs, social network analysis sentiment analysis, online, communicators, and listeners.

A theoretical and methodological comparison between personal influence and influencer logic

Full size image

What has been discussed so far lays the groundwork for the comparison announced at the beginning of this paper between the proposal of the Katz-Lazarsdelfian tradition of the notion of personal influence and the one of influencer logic. Starting from this comparison, the question is whether what is happening in the field of the analysis of the big data provided by the spread of the digital footprint is capable of adding some new element to what has already been highlighted by the “two-step communication theory”, or whether it simply represents its explication. Is something happening that Katz and Lazarsfeld predicted? Or do the analyses of networks and feelings even conflict or contradict the two authors?

Pre-existence of Networks and Co-construction of Networks

A first point for reflection is that the underlying logic is not dissimilar: the 1944 proposal placed at the centre of the process of influence that the media exerted on the population the figure of the opinion leaders who, through their personal influence, played a strategic role in the dissemination of opinions among opinion followers; today influencers exert their influence in the dissemination of opinions through digital channels by creating contacts with their followers. However, while the concept of “personal influence” presupposes the existence of networks and influence that pre-exist the communication flow, the concept of “influencer” seems to allude to the fact that nodes are, in some ways, created by the very use of communication, in a logic of co-construction of networks.

In this respect, we recall that unlike virtual methods (Hine, 2005), which involve adapting methodological strategies developed offline to digital environments, digital methods use the nature and affordances of online environments to understand how digital devices (search engines, social platforms, functions such as retweeting) structure communication and interaction flows on the Internet (Caliandro & Gandini, 2019, pp. 38–39). Close attention should be paid to the “structuring” concept. The inspiration comes from the well-known action-network theory of Bruno Latour (1988) and Michel Callon (1986), a theoretical and methodological approach that investigates the network of relations between human beings and material objects and the consequent co-construction of social reality resulting from their interaction. Digital users and devices, human and non-human actors, are “co-authors of social research, as they provide the researcher with the so-called ‘naturally digital methods’ necessary for the analysis of digital life forms and the emic categories for their interpretation” (Caliandro & Gandini, 2019, p. 39). From this perspective, social media represent such a fluid and dynamic reality that social formations are to be considered as a result of the activity of actors, rather than a starting point for analysis (Postill, 2016).

Small Groups and Nodes-Hubs

Of course, the nature of the contacts between the two compared situations is not the same, the former involving face-to-face contacts, the latter virtual contacts. In this regard, let us return for a moment to Katz and Lazarsfeld’s proposal, to recall that one of the central points to which they turn their attention in their model of communication flow and to which they also devote a paragraph in Chap. 2 was “The ‘rediscovery’ of the small group”, “highlighting how they reached the idea that primary interpersonal relations could be an important intervening variable in the process of mass communication” (Katz & Lazarsfeld, 1955, p. 17). The authors thus explore different contexts in which this rediscovery can be witnessed: the industrial context, the armed forces, the urban community context, specifying “the common elements in this pattern of rediscovery”.

Well, what does this have to do with influencers, if some claim that “social media makes us more lonely” (Turkle, 2011)? It should be remembered—as mentioned in the first part of the paper—that the epistemological perspective underlying digital sociology rejects the techno-deterministic perspective; from this standpoint, it is therefore the way in which people nowadays live their online relationships that makes the difference. The technological medium does not come between the interacting parties; it is used for and is part of the interaction itself.

This is not the place to provide a complete picture of SNA, but it is interesting to highlight which key concepts are relevant to the parallel we are presenting between “modernity and tradition”. Even those who have little knowledge of SNA know that when analysing networks we are dealing with actors (also called nodes) and links (between these actors). There are some concepts that highlight the connection between the SNA and the perspective of Katz and Lazarsfeld. Among the basic metrics used in network analysis, we find the concept of centrality and, specifically degree centrality, which can indicate how easily actors with another out-degree centrality can exchange information with others or quickly spread it to many; this is the case of highly influential nodes-actors (Veltri, 2012); betweenness centrality measures the degree to which a node is connected to other nodes that are not connected to each other, i.e. the degree to which a node acts as a bridge. Nodes-actors with a high value of this measure can have considerable influence within a network by virtue of the control they have over the passage of information between others (Veltri, 2021, pp. 135–137). Furthermore, among the types of networks, we identify the ego network, which is based on a model in which there is a subject (ego) and all the people with whom he or she has a social link (alter) who are arranged in a series of four or five inclusive groups (circles), depending on the strength of their social links (Veltri, 2021, pp. 138–139). Finally, among the properties of networks, based on the structure and model of the network, the so-called scale free network, characterised by the fact that the number of links from a given node, shows a distribution that follows the power law (power low distribution), which results in many actors-nodes with few links and few actors (the hubs) with a large number of links (see Barabási & Albert, 1999), hence, the ability of these hub actors to greatly influence other hubs because they are connected to so many people that the likelihood of them being connected to others who are easily persuaded is high (Veltri, 2021, pp. 142–144).

Survey, Content Analysis and Social Network Analysis and Sentiment Analysis

As with the previous point, from a technical-methodological standpoint, a change in the context and the way our lives and the interactions that take place in them are carried out is followed by an adaptation of the investigation tools and methods. Thus, whereas Katz and Lazarsfeld’s research used questionnaire surveys and content analysis, in social media social, we use network analysis and sentiment analysis.

From Offline to Online but with the Same Need to Belong and Assert One’s Identity

This is evident, for example, when it comes to consumption. Let’s try to think about the concept of fashion as it has changed over time. In the 1960s miniskirts were worn and were also a symbol of women’s emancipation, and in the 1970s the hippies chose a certain type of clothing as a sign of non-conformity; today there are punk and hipster styles. All these people share a claim to their freedom, a common need. Why should the 4.0 consumer be just a puppet at the mercy of good social marketers? Is it not the case that today, as in the past (if not even more so than in the past), the weakening of traditional ties has led to a landscape in which collectivity has been replaced by individuality but in the end the need to feel part of a community, to feel the strength of ties, the belonging to an identity has become urgent?

The search for an identity, a sense of belonging, leads to a desire for sharing, which often finds a response and space in the virtual community, which is a powerful tool for social representation. One answer—right or wrong—can be found on social channels and in the figures of influencers. Whether or not this is a bluff is not for us to determine. What needs to be emphasised is the fact that, both in sociology’s own digital methods and in our own lives, the transition from online to offline life is fluid. Therefore, also as digital social scientists, it is perhaps good to become aware of the changed theoretical and methodological frameworks, in order to be able to make (possibly critical) use of them.

From “Few Communicators and Many Listeners” to “All Communicators and All Listeners”

In the days of traditional mass media, most people were spectators and users, playing a passive role in receiving information, and just a few people had a voice. Today more people can speak (potentially everyone), and this also offers more freedom in choosing who to listen to. Therefore, it is not only the role of the communicators that is changed but also that of the listeners/followers.

This links to other crucial issues, such as power and minorities. Power is a privilege, and having access to the media (especially if it is widely distributed) is an opportunity for those who can speak out and a threat for those who would like to control communications in order to choose what is right to convey and what is not. If in the past there was only one story, today there are many stories. Among these stories there are more or less authoritative voices, more or less able to influence, but just as there is freedom to express oneself, there is also the freedom to choose who and what to listen to. That is why it is not so proper to use the word “influencer” with disdain, just because there are people who are more competent or more capable of using digital media to promote battles, themselves, products and why not to sell a powder pink or neon coloured nail polish worn by a man.