In this paper we study social aspects of using the Like button for purposes of impression management, identity construction, and maintenance of social ties online. On the theoretical level our investigation combines Goffman’s notion of face-work with concepts of social network analysis, shedding light on what we dub ‘nano-level’ interaction and sociality on social networking sites. Our data come from a 2013 classroom survey in which 26 Finnish university students were asked about their motives for and ways of using the Like button. Our results show that though the Like button was designed to allow users to express their positive evaluations of the contents of Facebook posts, comments, and pictures, it was in actual fact used for a wide variety of purposes, from dating efforts to conversation regulation and maintenance of social ties. Our results also reveal that the networked Facebook audience affects the users’ liking behavior, and that users reflect their liking based on previous likes.
1. Introduction: The importance of studying the Facebook Like button
2. A short history of the Like button
3. Theoretical background: Impression management in front of a networked audience
4. Interaction rituals in front of networked audiences
5. Empirical data and methods
6. Empirical results
1. Introduction: The importance of studying the Facebook Like button
With 1.44 billion monthly active users in early 2015 , Facebook is the leading social networking site in most countries worldwide (Kiss, 2014). For younger generations, Facebook is not only a tool or an application on the Internet but also a ubiquitous level of reality, where the boundaries between online and off-line actions are increasingly blurred (Wittkover, 2014). Especially for users of the Facebook mobile client, checking status updates and interacting with Facebook friends have become part of their daily routines.
Despite the growing literature on Facebook, there is a clear lack of research focusing on the use of the Like button . This research is, however, important for several reasons.
First, liking an object (post, comment, photo) is one of the most common, easily conducted, and simple forms of activity on Facebook. In May 2013, users liked an astonishing 4.5 billion objects every day . The Like button thus forms a common and, to our mind, crucial part of a particular type of online interaction with its own platform-specific constraints and possibilities.
Second, in economic terms, the Like button is a central element in the emerging reputation-based cluster of business models that power a growing share of Web-based companies. Gerlitz and Helmond (2013) claim that “the Like economy” is the third step in the evolution of Internet money-making principles. In the first phase, the value of a Web site was relative to the individual visitors (hits) it received. This “hit economy” (Rogers, 2002, cited in Gerlitz and Helmond, 2013) was then replaced by Google’s PageRank algorithm, which was based on the number and value of the hyperlinks created between Web sites rather than on individual hits on a Web site. During the third phase, the Like button has emerged as the source of valuation for Web pages, giving birth to the Like economy.
To sell effective advertising, Facebook must be able to target advertising as precisely as possible. As Kosinski, et al. (2013) have shown, data on liking behavior can reveal intimate information about users without them being aware of it. A list of the average American male Facebook user’s likes — which Facebook possesses by default — discloses his sexual orientation, religion, political party membership, and other private information with accuracy varying between 80 percent and 90 percent . Thus, whether users realize it or not, they give away lot of personal and salable information about themselves by using the Like button. Not only does Facebook know about our private affairs and inclinations, but it also sells this information to third parties .
Third, the Like button also has important political implications and uses: The pros and cons of liking are frequently evoked in discussions concerning social media and political participation. The ease of liking a Facebook social action page may, so goes the argument of the pessimist camp, alienate people from taking to the streets with banners and slogans (Kristofferson, et al., 2014). The optimists, however, emphasize the importance of social media in political participation. A Pew study (Rainie, et al., 2012) found, for example, that 38 percent of U.S. social network users have used social media to “like or promote material related to politics or social issues.” Consequently, there are reasons to believe that the social plug-ins such as the Like button may play a role in the dispersion of political Web pages and Internet memes and lower the threshold for online participation with off-line consequences. Regrettably, this debate has not been backed up by detailed empirical studies concerning the political effects of Like button use.
Despite the growing research on Facebook in general, and the debates on the economic and political significance of Like button use in particular, the social aspects of the use of this Facebook feature have not, to our knowledge, been extensively studied.
In this study, we thus focus on the social practices, motives, and norms affecting the use of the Like button. We will examine the most minute and fleeting moments of Facebook users’ behavior (clicking a single button to send a positive signal) and maintain that this activity is bounded, among other things, by ritual elements of social interaction reminiscent of those that govern face-to-face interactions (Goffman, 1967). We will address the networked nature of the Facebook users’ audience in particular, claiming that it contributes to the specificity of online interaction — and liking behavior — on Facebook. We will show, for example, how the totality of the audience consisting of a user’s Facebook friends can be divided into smaller subsets affecting her liking behavior. In the next section, we provide a short description of the Like button and the history of its development. In the third section, we present our theoretical starting point, which draws from Erving Goffman’s notions of face-work and impression management, and combine it with concepts from social network analysis (Scott, 2000). In the fourth section, we analyze a pilot survey of the liking behavior of 26 Finnish Facebook users, and we offer conclusions in the final section.
2. A short history of the Like button
The Facebook Help feature explains that clicking the Like button underneath a Facebook post “is an easy way to let someone know that you enjoy it, without leaving a comment. Just like a comment though, the fact that you liked the post is visible below it.”  The Like button — represented by a thumbs-up symbol — also appears on Web sites other than Facebook. Clicking the button enables users to share the page with their friends on Facebook. Currently, likes are always visible to at least some other users.
The first version of the Like button, then called the Awesome button, was developed in 2007. Facebook Chief Executive Officer (CEO) Mark Zuckerberg was concerned about the public nature of the feature, and chose not to launch it until 2009 . In 2010, Like buttons were added to individual comments on posts and other objects . Since then, almost all public activity on Facebook has been likable. Facebook also started including posts and links users had liked in their timelines, as part of the activity log. This changed liking from a relatively private action (from one user to another, visible only to the friends of the user whose object was liked) to a fundamentally public gesture.
In essence, the Like button contains a binary code: Either users like an object, or they remain indifferent. Facebook does not contain a Dislike button; it was tested as a private feature, but Facebook decided that it would discourage people from sharing . In the text that follows, we will show how users employ this binary code — pressing or not pressing a button — for multiple and fine-tuned purposes of sociality, and how this use is affected by the expectations of and pressures from a user’s personal network.
As this short history of the Like button shows, the social gestures available to the user and the algorithms used to govern the visibility of such gestures are constantly changing. Researching sociality on Facebook, or on any social networking site for that matter, is like a never-ending race: The technological boundaries move and evolve in constant interaction with corresponding features of sociality.
3. Theoretical background: Impression management in front of a networked audience
Face-work on Facebook
The Like button was designed to let other users know that you enjoyed their comment, post, or picture. Since using the Like button is an inherently social signal, in order to understand its nature and use we must look beyond the properties of individual users and liked objects. This means shifting focus to the interactions between users, and to the ritual elements that influence interaction. In this effort, we draw on the work of Erving Goffman who, probably more than any other social scientist, was occupied with the issues of micro-level daily social encounters, often of a very mundane nature, and of building and maintaining one’s “face” in public. We maintain that the ritual of liking someone’s new Facebook profile photo is similar to the ritual of greeting anyone who comes into your office — not mandatory in a strict sense, but it may have negative consequences if neglected.
According to Goffman , “Every person lives in a world of social encounters, involving him either in face-to-face or mediated contact with other participants.” For Goffman, these encounters form the basis for the analysis of social interaction. He conceptualizes encounters or situations as strips cut from an ongoing stream of events. In these strips, people behave in a certain way that usually aims at smooth and conflict-free continuation of the situation. The participant in a particular situation tends to act out what Goffman calls a line — that is, a “pattern of verbal and non-verbal acts by which he [the participant] expresses his view of the situation and through this his evaluation of the participants, especially himself” . Goffman calls face the positive social value one claims for oneself by taking a certain line. When someone “has” a face, he or she feels at ease and can behave in a relaxed manner. In our daily life encounters we are continuously building, maintaining, and saving each other’s faces in a process Goffman calls “face-work.” Face-work takes a lot of effort and one is always running the risk of losing one’s face. For Goffman, “To study face-saving is to study the traffic rules of social interaction” .
Similar to Goffman’s analysis of off-line interaction, when users are acting (liking, posting, commenting) on Facebook, they are simultaneously trying to build an idealized image or face of themselves.
A number of studies of Facebook have indeed been inspired by Goffman’s ideas (Bullingham and Vasconcelos, 2013; Lim, et al., 2012; Strano and Wattai Queen, 2012; Wittkower, 2014). However, Bernie Hogan (2010) is critical of applying this kind of Goffmanian scheme to online situations. In Hogan’s opinion, Goffman employs a theatrical metaphor of public performances where egos and alters are co-present in the same physical space. Hogan himself prefers the metaphor of exhibition to conceptualize the specificity of Facebook behavior. According to this metaphor, Facebook algorithms function as an exhibition “curator” who assists users in presenting their works to the public.
We agree with Hogan that one should be careful when seeking counterparts to online phenomena in the off-line world. Nevertheless, we maintain that the metaphor of exhibition downplays important aspects of social interaction among Facebook users. We assert that, despite the absence of a common physical context, many Goffmanian ideas regarding impression management and face-work can be applied to Facebook. We follow the work of Wittkower (2014), who argues that the specific nature of Facebook (real names, etc.) essentially forces people to correlate their online identities with their off-line identities.
“Stretched face” and asynchronous temporality on Facebook
Goffman  notes that interaction in talk and in physical settings tends to proceed in spurts, an interchange at a time, with lulls between the interchanges. On Facebook and other mediated settings, this tendency is taken to an extreme. Instead of seconds and minutes, the discussion threads can stretch for hours or days, and there is often a considerable lag between gestures in these interactions. The strip of activity blends with the continuous stream of the Facebook news feed and the timeline easily fades into history without the clear-cut distinction available in physical settings.
Despite this asynchronous nature of interaction on Facebook, actions by users follow each other in a linear sequence, also influencing other users’ behavior. Even if the immediateness of a situation is taken away, the temporality and its effect on participants remain. This temporality, however, is of a different type and changes the ritualistic nature of the interaction and the nature of face. A wall post, for example, can be liked either immediately or years after the original posting.
In Goffman’s work, though face is tied to immediate situations, it is also a cumulative phenomenon — some of it is carried over to ensuing situations. On Facebook and in other similar online settings, as individual situations dissolve into a longer stream, the boundaries of face change, and it becomes stretched over an extended period of time. We call this phenomenon “stretched face,” referring to feeling safe and at ease in interaction among the specific audience a user has gathered for himself or herself. The stretched face also implies the dangers and fragility of such situations: An originally harmless Like may, with new contextual information, develop into a face-threatening affair weeks or months after the original post. Wittkower (2014) assumes that face-work and identity creation are identical processes. Despite the longer timeframe, even the stretched face cannot be considered a synonym for identity. While identity implies a more stable and substantive essence, the stretched face is anchored in survival of the situation and in maintaining a positive assessment both of the user herself and others. These actions feed into identity creation.
The Like button as an example of nano-level interaction
Goffman is interested in micro-gestures and reactions delivered via glances, gestures, and positionings, whether intended or not. On Facebook, these micro-gestures are different: Comments, emojis, and pictures form a major part of interactions, but they are more analogous to speech acts and verbal statements than to glances and positionings. We propose the term nano-level interaction as an equivalent to the Goffmanian micro-gesture, and use it to denote the most minute and fleeting forms of interaction online, such as liking.
The asynchronous nature of Facebook allows for more consideration and planning of nano-level interactions than the immediate nature of a physical social situation. On Facebook, we can take as much time as we need to evaluate all the situations, consider the audience, and ponder possible outcomes of different gestures. Asynchronicity also changes the permanence of actions. If a coworker walks up to you in a hideous new shirt and you let your face show your opinion, it cannot be undone. In the online setting, there is always time to reconsider every action.
In the Facebook context, liking someone’s post — along with other means such as commenting — can be interpreted as a form of support for the user’s face-work. However, faces are also vulnerable, and one is always running the risk of losing one’s face. On Facebook this could happen, for example, to a left-wing politician liking a neo-Nazi page. In addition, also not liking a posted object could lead to embarrassment and losing face, if a user neglects, say, her significant other’s new profile picture.
These Goffmanian insights presented above must be combined with the fact that Facebook friends form a particular type of audience consisting of people connected through personal network ties — an aspect we will address in the next subsection.
4. Interaction rituals in front of networked audiences
Facebook users write, like, and share various objects in public in front of a crowd of friends, relatives, colleagues, etc., whom they have invited or accepted as their Facebook friends, and with whom they maintain various kinds of social relations through Facebook and otherwise. However, at the same time, Facebook users are not always aware that a discussion that has the intimacy of a private conversation might actually be visible not only to the Facebook friends of both participants but also to a much larger network of third-step contacts, depending on the privacy settings of both users. In the following, we further develop the effect these implied audiences have on the ritualistic elements of Like button use (Waters and Ackerman, 2011, Debatin, et al., 2009).
We maintain that users’ behavior is likely to be influenced and constrained by the opinions of their Facebook friends. More importantly, this influence is not only dyadic (between user A and her particular Facebook friend B) but also networked: A’s relationship with B may be affected by B’s relationship with another of A’s Facebook friends, C.
danah boyd (2010) referred to the audiences of social network sites with the general notion of networked publics:
Networked publics are publics that are restructured by networked technologies. As such, they are simultaneously (1) the space constructed through networked technologies and (2) the imagined collective that emerges as a result of the intersection of people, technology, and practice.
These networked publics on social network sites often contain — as implied in the very notion of network — dense cliques or communities typically consisting of family members and colleagues, for example. In the vocabulary of social network analysis, the networked audience of Facebook can thus be conceptualized as a personal or egocentric network. This notion refers to a social system anchored around a focal individual (ego) who is connected to her network members (alters), who, in their turn, are interlinked with one another . In sum, we call a user’s Facebook friends a personal network audience (PNA), a notion that emphasizes its nature as a particular type of networked public consisting of a number of clearly defined individuals. This PNA is the potential audience in front of which all actions on Facebook happen.
When studying liking activity, this personal network audience has to be investigated at a closer range. As we previously noted, the network of users who can see an action is not always clear. And even if it were clear, the subjective perception of this network might matter at least as much as the actual network.
Therefore, user A who likes user B’s post on Facebook without scanning who else has liked it performs a positive evaluation not only in front of the personal network audience (PNA) of all her Facebook friends but also in front of a more or less clearly imagined network audience (INA) she has subjectively constructed in her mind of the complete personal network audience. (cf., Lampinen, et al., 2009). This imagined network audience reflects the user’s awareness of both her Facebook friends and the visibility of her actions.
The construction of imagined audiences and the networked nature of the Facebook audience have been dealt with in previous literature (e.g., boyd, 2010; Ellison and boyd, 2013; Marwick and boyd, 2011). Marwick and boyd , in their investigation of imagined audiences on Twitter note, for example, that participants in every mediated conversation have a sense of audience, which is often imagined by an individual in order to present herself appropriately. The authors further remark that on social network sites profile owners have been shown to be attentive to audience, and pay attention to ‘context collapse’, that is, the tendency of mainstream media technologies to bring together distinct audiences which in an off-line context would be separated .
Our own approach is distinct, however, in the sense that our analysis concerns liking behavior on Facebook, taking as its starting point the notion of a personal or egocentric network and its subsets, and their role in the user’s impression management efforts.
However, if A checks — before liking B’s post — who else has liked this post, user A will observe still a third audience, the network of previous likers (NPL), consisting of a network of specific, named individuals. The imagined network audience (INA) constructed subjectively by the user as well as the network of previous likers (NPL) form subsets of the complete personal network audience (PNA) .
Figure 1 illustrates the factors and audiences affecting liking behavior on Facebook. The symbols F1–F4 in the figure indicate the posting user’s Facebook friends, the undirected lines friendship relations within Facebook, and the lines with arrows the liking behavior of users. In other words, F1, F2, F3 and F4 are all Facebook friends of the posting user, F1, F2 and F3 are all Facebook friends with each other, but none of these three is Facebook friend with F4. Of these four actors, F1, F2 and F3 have liked the original post by the user.
Figure 1: The potential relationships (lines) between a user posting on Facebook and her friends (F1–F4), and liking of an object (arrows).
The figure suggests, first, that the relationship with the original poster of an object may have an impact on likes: We are more prone to like a post by a close Facebook friend than one by an acquaintance whom we have accepted as our friend somewhat reluctantly (Egebark and Ekström, 2011). Second, the quality, number, and network structure of previous likers are likely to affect one’s likes. This is probably even truer in the case of a sensitive or contradictory topic (e.g., a post on a political issue). Thus, if F1, F2 and F3 are close friends, F3 is more prone to like a post of controversial nature if F1 and F2 have both already liked it. Third, the imagined audience constructed subjectively by the user of the pool of all Facebook friends (some subset of F1–F4) is likely to influence liking behavior. These ideas are supported by previous research; Egebark and Ekström (2011) found that the number and quality of Facebook friends who had previously liked a post had an impact on users’ likes.
In sum, the analysis of Facebook liking behavior must simultaneously consider the rituals involved in public social interactions such as liking and the networked nature of the various audiences which are likely to affect the users’ likes. Face-work in front of the networked audiences combines both of these aspects: Face-work in online situations is the asynchronous monitoring of (liking) behavior attempting to maintain a positive valuation from all participants in the situation. Since the networked online situation stretches from immediate reactions to the resurfacing of past actions, the face has to stretch as well, and take into account its imagined network audience. In the next section, we present our methods and empirical data.
5. Empirical data and methods
In this section, we discuss the results of a classroom pilot survey conducted in the spring of 2013 among 26 Finnish university students. We chose a convenience sample of students due to the exploratory nature of this article, and our focus on understanding the broader framework of sociality on Facebook. The respondents were 24.5 years old on average, seven were male, and they had had a Facebook account for an average of five years. They were given a structured questionnaire concerning their motives and methods for using the Like button (see Appendix). At the beginning of the questionnaire, they were instructed to reflect as thoroughly as possible on their use of the button.
The first group of questions concerned social pressures. Based on our theoretical views presented in the previous section we hypothesized that, in order to conduct face-work on Facebook, the users pay constant attention to what other users are thinking. We thus inquired if the respondents in general considered the opinion of their Facebook friends when they used the Like button (personal network audience, PNA), and if so, if they had a specific friend or friends in mind (imagined network audience, INA). We also asked if the respondents checked, before liking an object in Facebook, who else had liked it (network of previous likers, NPL); if they had had regrets regarding their liking activities because of social reasons (face-work); and if they had tried to align their likings with their Facebook friends’ perceptions of their Facebook behavior (face-work). For each question, we inquired how frequently this specific aspect had affected their liking (the options were never, sometimes, often, always, can’t say).
The second group of questions addressed the different types of liking on Facebook. We asked these questions to gain insight into multiple motives and uses of the Like button in social interaction. In addition to the Like button’s intended use as a positive evaluation of other users’ posts, we wanted to find out how the Like button was used to build up and maintain a public face and to make, strengthen, and break relationships. We compiled a list of 16 ways or modes of using the Like button and asked, similarly to the first group of questions, about the frequency each mode’s application. The first and seemingly most natural mode on the list was called genuine like. This referred to a mode of liking in which the user liked the contents of the object (e.g., a nice photograph of a kitten). In addition, however, our list of types of liking contained 15 other items ranging from pity like (“I liked the object because I felt pity toward the poster”) to the use of the Like button for dating purposes. Finally, we asked respondents to describe in their own words other possible reasons for using the Like button.
Since 26 Finnish university students comprise a small and a highly selected convenience sample, the data cannot be considered statistically representative. Moreover, national and cultural differences may yield a further bias on the results. Nevertheless, we think this pilot study yields interesting information on a topic that has thus far been rarely investigated. In addition, it will benefit future research on the use of the Like button and other social plug-ins.
6. Empirical results
Impact of the network of previous likers and imagined audiences on likes
Our first results show that Facebook users do at least sometimes think of the social impact and consequences of their liking behavior both prior to and after using the Like button. Table 1 reveals that only five out of 26 students chose the option never when they were asked: “When you press the Like button in Facebook, do you consider what your Facebook friends think of your liking,” while 21 chose either the option sometimes or often. Similarly, 15 out of 26 respondents had aligned their liking behavior according to the expectations of their Facebook friends, and 16 had constructed an imagined networked audience at least sometimes for their liking behavior (Table 1, rows 1–3). In a like manner, 23 respondents had sometimes checked the network of previous likers and ten had sometimes worried about losing face because of their liking behaviors (Table 1, rows 1–3).
Table 1: Social pressures affecting Facebook likes.
Note: The options given to respondents were 1=never, 2=sometimes, 3=often, 4=always, 5=can’t say. In the table options 2, 3, and 4 have been merged in the category ‘at least sometimes’.
never at least sometimes can’t say total When you press the Like button, do you consider what your FB friends will think of your liking? 5 21 0 26 Has the fact that you know that your liking fits into your FB friends’ perception of your FB behavior influenced your liking? 9 15 2 26 When you consider your FB friends’ opinions, do you have some specific person/s in mind? 9 16 1 26 Before liking an object on FB, do you check who else has liked it? 3 23 0 26 Have you regretted any of your likes due to social reasons? 16 10 0 26
The results in Table 1 suggest that social pressures are involved in users’ liking activities at least sometimes. It is reasonable to presume that this tendency is even more elevated in the case of sensitive postings.
Table 2: The frequency of the factors that affect liking.
Note: The categories are not mutually exclusive. Legend of frequency: 1=never, 2=sometimes, 3=often, 4=always, 5=can’t say)
Name Explanation Frequency N 1 2 3 4 5 Content like I liked the contents of the object 0 2 11 13 0 26 Pity like I liked the object because no one had liked it, and I felt pity for the user who posted it 17 9 0 0 0 26 Network like I checked who else had liked the object, and they included many people important to me 11 13 2 0 0 26 Friendship like The user who posted the object is an important person to me 2 12 10 2 0 26 Reciprocal like I liked the object posted by user A because she usually likes my posts 11 13 1 0 1 26 Façade like I liked the object because I want to look good in the eyes of my FB friends 13 12 0 1 0 26 Routine like Whenever I use Facebook, I routinely go through the updates of some friends and like them 19 5 2 0 0 26 I noted your post like I liked an object in order to show that I had noticed it 1 11 12 2 0 26 Welcome like I liked the object to welcome a new Facebook user 15 5 5 0 1 26 Tie maintenance like I liked the object to maintain the relationship between myself and my FB friends 8 14 4 0 0 26 Dating like I liked the object to flirt with the user who posted it 19 7 0 0 0 26 Newsfeed like I liked the object to make sure I keep getting this user’s updates in my newsfeed 20 4 1 0 1 26 Pressure like My Facebook friend sent me a private message and pressured me to like the object 21 5 0 0 0 26 Instrumental like I liked the object because the user who posted it might be useful to me in the future 14 10 0 0 2 26 Joke like I liked the object as a joke — nobody expects me to like posts like this! 16 9 0 1 0 26 Harassment like I liked the object to harass the user 25 1 0 0 0 26
Table 2 suggests, not unexpectedly, that the contents of the liked object have the most important influence on liking. However, half of the respondents chose either the option often or sometimes (instead of always) even for genuine like. In addition to content, other factors thus also play a role in the use of the Like button.
The two other most popular types of likes were friendship like (user liked the object because the poster was an important person to him or her) and I noted your post like (user liked the object posted on his or her Facebook wall to convey the message “I noticed your post, thank you for it”). These actions, in addition to the positive signal, carry meanings related to the stability of the relationship between the liker and the receiver of the like: They are intended to signal strengthening of the connection. The I noted your post like is a direct example of face-work: Because of the asynchronous nature of Facebook, one can never be sure whether another user has noticed a wall post or a certain uploaded picture. The Like button is used as the minimal, positive reaction, acknowledging the post and the relationship between the users. Without this kind of acknowledgment, the situation could be perceived as rude and face-threatening.
These findings speak of the important impact of the social effect on liking behavior. They were further supported by the fact that almost half of the respondents had at least sometimes used the options network like, façade like, and tie maintenance like. With the network like (referring to the network of previous likers, NPL), the user liked the object because many other Facebook friends who were important to him or her had already liked it. With the façade like, the user liked the object because he or she wanted to look good in the eyes of his or her Facebook friends (imagined network audience, INA). The façade like in particular is a good example of face-work on Facebook: Users evaluate what other users think about them, and alter their behavior accordingly. Finally, with the tie maintenance like, the user liked the object because he or she wanted to maintain the relationship with his or her Facebook friend.
Multiple practices, motives, and meanings attached to the Like button
We also asked the respondents to add to our premade list their own suggestions concerning the various types of likes. As a result, we got an astonishing number and variation of new ways to use the Like button. Examples included a letting steam out like (meant to make the liker himself or herself feel good); an I am alive like (informing others that the user is following debates though he or she does not participate actively); a profile picture like (in Facebook, it is considered polite to like new profile pictures. Not liking the new profile picture of one’s close friend would threaten her face); a conversation regulation like (the Like button was used to notify that a comment was read, but a combination of liking and not giving a textual response was used to signal that the conversation had reached its terminal point), and even not liking (the user demonstratively does not like an object everyone expects him or her to like).
Superficially simple in design, the Like button thus functions — instead of as a binary code only — as a prism for very sophisticated and multiple forms of social interaction. The number of different uses suggested by our respondents implies a complex web of social interactions all woven together via a single, binary medium. Reminiscent of irony, sometimes only a close circle of friends can recognize all the complex meanings embedded in a single like. For example, in order to understand the manner of using the Like button to signal one’s Facebook friends that the user still reads Facebook even though she is not posting at all (I am alive like), one must have been keeping an eye on the actions of this particular user. In other cases, users have developed more general Facebook behavior etiquette and norms for liking, such as the norm of liking an object posted on one’s Facebook wall.
In sum, the open-ended descriptions of the liking activities reveal the users’ imaginative abilities to use a single button to develop, maintain, and end social relationships and conversations, to get involved in social exchanges, to balance between sometimes contradictory expectations of their audience, and to build and maintain a face in front of their networked audiences. Among the multitude of different uses, many could be interpreted as ways of regulating and setting the temporal boundaries of interaction in a non-corporeal context. These uses include regulating conversation via the Like button, maintaining a façade, and signaling that one is reading comments, even though not participating in the actual debate.
In this paper we have tried to shed light on the social factors affecting the use of the Like button on Facebook. To conceptualize and make sense of the interaction on Facebook we combined Goffmanian analysis with the egocentric network approach. Our empirical data suggest that users do reflect their liking based on previous likes, and that the networked audience affects the users’ liking behavior. We also found that liking is used in a wide variety of ways ranging from regulating conversation to signaling the strength of the tie between users and maintaining a face.
Our results support the results of previous studies on the effect of Facebook use on sociality. Karakayali and Kilic  claimed, for example, that with the increasing amount of time spent on Facebook users may become overly interested in and dependent upon the opinions and behavior of their Facebook friends. Consequently, users on social networking sites may tend to become highly inquisitive and observant about their network and social ties leading to an enhanced form of network consciousness. This growing network sensitivity, the authors propose, might even spread outside Facebook.
Closer to the purposes of the current research, our investigation is in line with the only study of peer pressure related to liking behavior on Facebook we are aware of. A recent study conducted by two economists found “herding behavior” among users and identified conformity as the main mechanism (Egebark and Ekström, 2011). Although in agreement with these findings, our study suggests that other, more versatile and complex mechanisms could be detected.
Our theoretical elaboration of face-work and interaction rituals in front of a networked audience helps to explain these previous results. The herding behavior and peer pressure operate through face-work which aims at supporting and saving faces. When users try to maintain a positive image of themselves in front of networked audiences, at least some deviations from norms are being avoided. The same factors (e.g., wanting to maintain a positive face) that regulate our interactions in physical settings thus also affect the way we act on online social networking sites. The minutiae of these rituals may be different online, but their spirit remains the same.
In this article we have addressed the temporality of social interaction on Facebook. We maintain that the stretched nature of online interactions leads to stretched face, and to a situation where past actions may resurface in a different context. The wider effects of this asynchronicity form an interesting area for future study.
Although we have concentrated exclusively on Facebook, similar social pressures might exist in other online social media. YouTube has comments and upvotes, Twitter favorites and retweets, and even many discussion forums and comment sections of newspapers have upvoting features. Due to the difference in the basic construction and different functionalities of YouTube, Twitter, and Facebook, contextualized case studies are needed, instead of lumping all applications into the vague category of social media. The Facebook Like button, however, is an example of how users find new, innovative, and versatile means of social interaction in online media, and how these interactions are affected by social pressures. While some phenomena of online interaction are truly novel, others can be understood by applying the accumulated knowledge from social science studies concerning offline interaction situations.
About the authors
Veikko Eranti is a Ph.D. student at the Department of Social Sciences at the University of Helsinki.
E-mail: veikko [dot] eranti [at] helsinki [dot] fi
Markku Lonkila is a professor of sociology at the University of Jyväskylä.
E-mail: markku [dot] lonkila [at] jyu [dot] fi
Eranti acknowledges a working grant from the Kone Foundation. Lonkila acknowledges the support of the Finnish Centre of Excellence “Choices of Russian Modernisation.”
1. http://venturebeat.com/2015/04/22/facebook-passes-1-44b-monthly-active-users-1-25b-mobile-users-and-936-million-daily-users/, accessed 16 May 2015.
2. The huge popularity of the platform has given birth to a burgeoning literature on Facebook by computer, social, and political scientists alike (e.g., Bernstein, et al., 2013; Birnbaum, 2008; Bullingham and Vasconcelos, 2013; Conroy, et al., 2012; Lampinen, et al., 2009; Vitak, et al., 2012). For a recent review, see Wilson, et al. (2012). However, the bulk of the writing on the Like button consists either of non-academic journal articles or examines the role of the button in marketing management (e.g., Schöndienst, et al., 2012). A rare exception to the rule is a study by two economists who, based on a natural experiment through five users’ Facebook accounts, investigated the users’ liking behavior in 2010. The researchers sent a Facebook post through the five users’ accounts and studied whether the probability of it being liked by, say, the users’ Facebook friend A was affected by this post having already been liked by a) one user unknown to A, b) three unknown users, or c) one of As Facebook friends. The two latter cases were confirmed by the study, and the authors interpreted the results through the conformity effect (Egebark and Ekström, 2011).
3. https://www.facebook.com/photo.php?fbid=10151908376831729, accessed 20 August 2014.
4. Male homosexuality could be predicted with 88 percent, religion with 82 percent, and political party membership with 85 percent accuracy (Kosinski, et al., 2013).
5. https://www.facebook.com/business/power-of-advertising, accessed 20 August 2014.
6. https://www.facebook.com/help/like, accessed 20 August 2014.
7. http://www.quora.com/Facebook-Inc-company/Whats-the-history-of-the-Awesome-Button-that-eventually-became-the-Like-button-on-Facebook, accessed 20 August 2014.
8. https://www.facebook.com/blog/blog.php?post=399440987130, accessed 20 August 2014.
9. http://www.quora.com/Facebook-1/Why-is-there-no-Dislike-button, accessed 20 August 2014.
10. Goffman, 1967, p. 5.
12. Goffman, 1967, p. 12.
13. Goffman, 1967, p. 37.
14. For the basics of social network analysis, see Scott (2000).
15. Marwick and boyd, 2011, p. 115.
17. Conceptualizing Facebook in this way opens up the possibility of combining the most relevant features of the Goffmanian impression management perspective with the tools of social network analysis. Since in this study we could not collect data on the network ties between the respondents’ Facebook friends, we operated on the level of the personal network audience, imagined audience, and network of previous likers. The detailed effect of the network structures and cliques on likes, for example, remains an important task for future research. It will be facilitated by the development of specialized applications with which to visualize and measure Facebook networks.
18. Karakayali and Kilic, 2013, p. 175.
19. Swarts, 2009, p. 284.
20. Swarts, 2009, p. 284.
Michael S. Bernstein, Eytan Bakshy, Moira Burke, and Brian Karrer, 2013. “Quantifying the invisible audience in social networks,” CHI ’13: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 21–30.
doi: http://dx.doi.org/10.1145/2470654.2470658, accessed 27 May 2015.
Mathew Birnbaum, 2008. “Taking Goffman on a tour of Facebook: College students and the presentation of self in a mediated digital environment,” Ph.D. dissertation, University of Arizona, at http://arizona.openrepository.com/arizona/bitstream/10150/194670/1/azu_etd_2833_sip1_m.pdf, accessed 27 May 2015.
danah boyd, 2010. “Social network sites as networked publics: Affordances, dynamics, and implications,” In: Zizi Papacharissi (editor). Networked self: Identity, community, and culture on social network sites. New York: Routledge, pp. 39–58.
Liam Bullingham and Ana C. Vasconcelos, 2013. “‘The presentation of self in the online world’: Goffman and the study of online identities,” Journal of Information Science, volume 39, number 1, pp. 101–112.
doi: http://dx.doi.org/10.1177/0165551512470051, accessed 27 May 2015.
Meredith Conroy, Jessica T. Feezell, and Mario Guerrero, 2012. “Facebook and political engagement: A study of online political group membership and offline political engagement,” Computers in Human Behavior, volume 28, number 5, pp. 1,535–1,546.
doi: http://dx.doi.org/10.1016/j.chb.2012.03.012, accessed 27 May 2015.
Bernhard Debatin, Jennette P. Lovejoy, Ann-Kathrin Horn, and Brittany N. Hughes, 2009. “Facebook and online privacy: Attitudes, behaviors, and unintended consequences,” Journal of Computer-Mediated Communication, volume 15, number 1, pp. 83–108.
doi: http://dx.doi.org/10.1111/j.1083-6101.2009.01494.x, accessed 27 May 2015.
Johan Egebark and Mathias Ekström, 2011. “Like what you like or like what others like? Conformity and peer effects on Facebook,” Research Institute of Industrial Economics, Working Paper, number 886 (14 October), at http://www.ifn.se/wfiles/wp/wp886.pdf, accessed 27 May 2015.
Nicole Ellison and danah boyd, 2013. “Sociality through social network sites,” In: William H. Dutton (editor) Oxford handbook of Internet studies. Oxford: Oxford University Press, pp. 151–172.
doi: http://dx.doi.org/10.1093/oxfordhb/9780199589074.013.0008, accessed 27 May 2015.
Carolin Gerlitz and Anne Helmond, 2013. “The Like economy: Social buttons and the data-intensive Web,” New Media & Society, volume 15, number 8, pp. 1,348–1,365.
doi: http://dx.doi.org/10.1177/1461444812472322, accessed 27 May 2015.
Erving Goffman, 1967. Interaction ritual; Essays on face-to-face behavior. Garden City, N.Y.: Anchor Books.
Bernie Hogan, 2010. “The presentation of self in the age of social media: Distinguishing performances and exhibitions online,” Bulletin of Science, Technology & Society, volume 30, number 6, pp. 377–386.
doi: http://dx.doi.org/10.1177/0270467610385893, accessed 27 May 2015.
Nedim Karakayali and AYsenur Kilic, 2013. “More network conscious than ever? Challenges, strategies, and analytic labor of users in the Facebook environment,” Journal of Computer-Mediated Communication, volume 18, number 2, pp. 61–79.
doi: http://dx.doi.org/10.1111/jcc4.12005, accessed 27 May 2015.
Jemima Kiss, 2014. “Facebook’s 10th birthday: From college dorm to 1.23 billion users,” Guardian (4 February), at http://www.theguardian.com/technology/2014/feb/04/facebook-10-years-mark-zuckerberg, accessed 20 August 2014.
Michal Kosinski, David Stillwell, and Thore Graepel, 2013. “Private traits and attributes are predictable from digital records of human behavior,” Proceedings of the National Academy of Sciences, volume 110, number 15 (9 April), pp. 5,802–5,805.
doi: http://dx.doi.org/10.1073/pnas.1218772110, accessed 27 May 2015.
Kirk Kristofferson, Katherine White, and John Peloza, 2014. “The nature of slacktivism: How the social observability of an initial act of token support affects subsequent prosocial action,” Journal of Consumer Research, volume 40, number 6, pp. 1,149–1,166.
doi: http://dx.doi.org/10.1086/674137, accessed 27 May 2015.
Airi Lampinen, Sakari Tamminen, and Antti Oulasvirta, 2009. “All my people right here, right now: Management of group co-presence on a social networking site,” GROUP ’09: Proceedings of the ACM 2009 International Conference on Supporting Group Work, pp. 281–290.
doi: http://dx.doi.org/10.1145/1531674.1531717, accessed 27 May 2015.
Sun Sun Lim, Shobha Vadrevu, Yoke Hian Chan and Iccha Basnyat, 2012. “Facework on Facebook: The online publicness of juvenile delinquents and youths-at-risk,” Journal of Broadcasting & Electronic Media, volume 56, number 3, pp 346–361.
doi: http://dx.doi.org/10.1080/08838151.2012.705198, accessed 27 May 2015.
Alice Marwick and danah boyd, 2011. “I tweet honestly, I tweet passionately: Twitter users, context collapse, and the imagined audience,” New Media & Society, volume 13, number 1, pp. 114–133.
doi: http://dx.doi.org/10.1177/1461444810365313, accessed 27 May 2015.
Lee Rainie, Aaron Smith, Kay Lehman Schlotzman, Henry Bardy, and Sidney Verba, 2012. “Social media and political engagement,” Pew Research Center (19 October), at http://www.pewinternet.org/2012/10/19/social-media-and-political-engagement/, accessed 27 May 2015.
Valentin Schöndienst, Florian Kulzer, and Oliver Günther, 2012. “Like versus dislike: How Facebook’s like-button influences people’s perception of product and service quality,” ICIS 2012 Proceedings, at http://aisel.aisnet.org/icis2012/proceedings/DigitalInnovation/4/, accessed 27 May 2015.
John Scott, 2000. Social network analysis: A handbook. Second edition. London: Sage.
Michele M. Strano and Jill Wattai Queen,2012. “Covering your face on Facebook: Suppression as identity management,” Journal of Media Psychology, volume 24, number 4, pp. 166–180.
doi: http://dx.doi.org/10.1027/1864-1105/a000076, accessed 27 May 2015.
Jessica Vitak, Cliff Lampe, Rebecca Gray, and Nicole B. Ellison, 2012. “‘Why won’t you be my Facebook friend?’ Strategies for managing context collapse in the workplace,” iConference ’12: Proceedings of the 2012 iConference, pp. 555–557.
doi: http://dx.doi.org/10.1145/2132176.2132286, accessed 27 May 2015.
Susan Waters and James Ackerman, 2011. “Exploring privacy management on Facebook: Motivations and perceived consequences of voluntary disclosure,” Journal of Computer-Mediated Communication, volume 17, number 1, pp. 101–115.
doi: http://dx.doi.org/10.1111/j.1083-6101.2011.01559.x, accessed 27 May 2015.
Robert Wilson, Sam Gosling, and Lindsay Graham, 2012. “A review of Facebook research in the social sciences,” Perspectives on Psychological Science, volume 7, number 3, pp. 203–220.
doi: http://dx.doi.org/10.1177/1745691612442904, accessed 27 May 2015.
D.E. Wittkower, 2014. “Facebook and dramauthentic identity: A post-Goffmanian theory of identity performance on SNS,” First Monday, volume 19, number 4, at http://firstmonday.org/article/view/4858/3875, accessed 27 May 2015.
doi: http://dx.doi.org/10.5210/fm.v19i4.4858, accessed 27 May 2015.
Received 20 August 2014; revised 25 May 2015; accepted 26 May 2015.
This paper is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
The social significance of the Facebook Like button
by Veikko Eranti and Markku Lonkila.
First Monday, Volume 20, Number 6 - 1 June 2015