Facebook and Twitter have been studied from different perspectives and within different contexts for over a decade. These two social media platforms are among the world’s most frequently used [1
]. They have thus attracted the attention of various business, political, and public administration organizations, as well as NGOs (Non-governmental Organizations) and NPOs (Nonprofit Organizations) (e.g., [2
]). These platforms are primarily text-based. Previous social media content research has therefore mainly focused on written words. This is, however, changing with the growth of social media platforms that predominantly center on images. Instagram is at the forefront of such platforms, having surpassed Twitter in terms of number of users and time spent on the platform [8
]. Furthermore, the use and influence of images, pictures, and videos are constantly increasing on social media—also on (to date) largely text-based social media platforms, such as Facebook (e.g., [10
The image-sharing service Instagram was launched in October 2010 and has in the interim gained increasingly in popularity. In December 2014, Instagram had over 300 million users worldwide [12
] and, in September 2015, 400 million people worldwide [13
] already used the platform to share photos and videos. The instant sharing of visual moments of the user’s everyday life has made the platform popular. There is reason to believe that online communication processes change when centered on visuals [14
]. Hence, we believe, scholars have to add Instagram to the social media ecology when studying online communication.
In the spring of 2016, we conducted a literature search of the largest social science databases from 2010 onwards (the word “Instagram” had to appear in the abstract or among the keywords) and it shows that, to date, there is very little research available with respect to the Instagram platform. The majority of these few studies focuses on the individual user and investigates aspects such as self-presentation and self-imaginary (e.g., [15
]). The use of Instagram from an organizational perspective is very seldom discussed [17
]. We refer to a few of these studies in more detail throughout the article.
At this early stage of Instagram research, the question arises: How should this platform, centered on images rather than text, be studied? Specifically, we are interested in organizations’ use of Instagram in order to investigate this platform’s strategic application. We initially started researching Instagram by studying Swedish political parties’ use of it during the 2014 national elections [18
]. Previous studies on political communication conducted with respect to other social media platforms, notably Facebook and Twitter, inspired us to study an election campaign. We focused on whether parties use Instagram to disseminate information, foster participation, and/or to allow its users to engage in discussions. We also considered research from the field of visual communication in our Instagram study when discussing a methodology to analyze organizations’ use of the platform. Is Instagram only used to share one-way information, focusing on spreading information and self-presentation? Or is it used for two-way communication to establish and cultivate organization-public relationships? These questions can also be applied to other types of organizations adopting this platform.
The main findings of our studies of Swedish political parties are that they primarily use their official Instagram pages to disseminate information. Instagram pictures (including their captions) were rarely used to mobilize supporters. Most pictures included one (or more) individual(s), with top candidates appearing very often, primarily in a political/professional context [19
]. Similar to previous research on Facebook and Twitter, Swedish parties hardly used Instagram to interact with their followers in order to foster organization-public relationships [18
In this article, we present and briefly discuss our methodological framework to study Instagram, hence contributing to the research on organizations’ use of social media platforms and of online visual communication. Our hope is that the readers of this Special Issue will use as well as adapt our methodological framework to advance research on Instagram or other image-based social media platforms. Essentially, the main focus is on the applicability of the variables (perspective, broadcasting, mobilization, personalization, privatization, celebrities, hybridity, shared content, campaign references, and interactivity) and not on their ontology and theoretical background.
In the next section, we provide a brief introduction to online visual communication, with a particular focus on organizations’ use of visual communication for information purposes and with respect to managing organization-public relationships. The main part of this article then follows with an introduction and explanation of the variables we used in the 2014 Swedish study, as well as their broad application in the study of organizations’ use of Instagram.
Online Visual Communication: An Emerging Field
Visual communication is a rather new field within communication sciences [20
]. Scholars describe it as an “emerging field” [21
]. It has gained importance in recent years, as visual media has significantly impacted our daily lives in terms of the way information is disseminated and used. However, current research has hardly considered online visual communication [22
]. Theories of (online) visual communication and methodological approaches as well as empirically-based studies [20
] are still needed. Müller [22
] (p. 24) defines visual communication research in the 21st century as:
“[A]n expanding subfield of communication science that uses social scientific methods to explain the production, distribution and reception processes, but also the meanings of mass-mediated visuals in contemporary social, cultural, economic and political contexts. Following an empirical, social scientific tradition that is based on a multidisciplinary background, visual communication research is problem-oriented, critical in its method, and pedagogical intentions, and aimed at understanding and explaining current visual phenomena and their implications for the immediate future.”
This understanding of visual communication forms our point of departure in this article.
Visual communication research is often based on what could be described as an iconic turn [24
]. In this context, images are understood as an additional source of communication complementing written or spoken text. Images suggest reality and thus have the potential to build and/or restore trust and legitimacy with respect to the sender (i.e., organizations). Images help the viewer to achieve coherence faster; they create meaning and foster interaction [24
Previous studies in the field of visual communication have highlighted that pictures draw viewers’ attention more effectively than text [26
] and they “have a powerful impact by increasing viewers’ attention and retention” [29
] (p. 133). For instance, in marketing, the importance of visual elements has increased constantly over the past decades. Examining changes in magazine advertisements’ style between 1969 and 2002, McQuarrie and Philips [30
] note that, over time, more weight has been given to pictorial elements and less to verbal elements. “If we ask why the style of advertising has changed, it appears that while the older ads assume an attentive reader
, the more recent ads presume a visually oriented, casually browsing viewer
] (p. 96; emphasis in original text). Examining Instagram with regard to its use in communication in politics, businesses, NGOs, etc., thus seems to be of great importance. Analytical approaches to images (in political communication and marketing) show that, with regard to viewers, images “can have rhetorical impact and make persuasive arguments” [14
] (p. 122). Images create causal relationships and thereby influence how “individuals view (…) products and services” [29
] (p. 122). Images may have an agenda-setting function, may dramatize policy, help with emotional appeals, help build a candidate’s or brand’s image, help create identification, and help viewers connect to societal symbols [14
]. Images that also use text may communicate messages even more effectively [14
], as the text may provide additional and more detailed information [31
]. Based on this past research, we expect communication processes to change substantially when a platform focusing on images rather than on text—such as Instagram—enters the ecology of online communication platforms. Studying it is therefore important.
2. Methodological Framework: How to Analyze Instagram?
First of all, what is Instagram? Instagram is a platform centered on uploading and sharing pictures and videos that may or may not be textually tagged with a caption on which followers may or may not comment. On Instagram, user roles are divided into posters and followers. A poster is the owner of an Instagram account and only posters can publish a new picture or video on their accounts. A follower is someone who subscribes to a poster’s account and then receives pictures and videos (updates). We distinguish between postings and posts: A posting is only the information—the picture and video with or without a caption—that the poster provides. A post also includes followers’ reactions to the posting, such as comments and likes.
In our previous studies, we primarily focused on political parties’ postings [19
] and activities in the comment field [18
]. Since we started studying Instagram in 2014, the platform has introduced new features. Today, followers may also “like” a posting by publishing a heart-shaped icon and can re-post a posting in their own feed by simply pressing an arrow shaped icon underneath the posting.
In this article, we present the variables we have developed and used to study Instagram, and also draw on previous studies of Facebook and Twitter. Figure 1
provides an overview of the variables of our methodological framework, which we describe in detail below. We grouped the variables into four clusters: perception, image management, integration, and interactivity. We present each cluster and its variables in the following. We have, to date, only used these variables when studying political organizations’ (i.e., parties) use of Instagram, but argue that these variables may also be used for studying other types of organizations and their use of the platform.
The content analysis is set out to account for the “how” of social media: how organizations use Instagram to reach out to their stakeholders. Hence, when coding, we took the follower’s perspective in order to code how the stakeholder perceives an organization’s Instagram communication.
We downloaded and archived the material to code the posts on Instagram and in order to revert to the material at a later stage.
First, we directed our attention to the posting (the picture/video with or without its caption) and we asked: How is the posting perceived when the follower views it for the first time? In this regard, we were specifically inspired by research stating that social media are primarily used for strategic broadcasting (spreading stances on current issues), as well as for mobilization (convincing supporters to vote and to take part in a demonstration, motivating consumers to buy a product, etc.). Visual communication studies also show that images are particularly important for creating relationships with regard to how followers view an organization. We therefore also coded for perspective, i.e., whether the posting can be perceived as an official statement or simply a snapshot view of the organization.
Pictures published on Instagram are mostly taken with a mobile device, such as smartphones and tablets. These photos enable users to share visual snapshots of their everyday life. However, it raises the question: Does the organization use this affordance or does the organization want to portray itself in an official manner/context? That is, postings can be made to look professional or casual, selfie-like. Therefore, we coded from the perspective from which the posting was perceived: Was it perceived as an official picture of an organization/organizational actor (e.g., the top candidate of a political party or the CEO of a company) or rather as a snapshot? The posting is thus coded as reflecting a (rather) official context or a (rather) snapshot/selfie context and, if neither applies, coded as being not applicable.
Previous studies on political parties’ use of social media (e.g., [32
]), and that of candidates/politicians (e.g., [34
]), companies [2
], and NPOs [7
], show that Facebook, Twitter, and YouTube are mainly used to spread an organization’s stances on current issues, opinions, and information on services and products to the community. Social media certainly allow organizations to bypass traditional media in order to broadcast their choice of information. Hence, the question arises: Is Instagram used to broadcast information across interlinked personal networks? In this regard, visual communication studies also underline the agenda-setting function of images and that images are very effective in communicating messages [14
], specifically in combination with text (captions may provide additional and/or more detailed information) [31
The variable broadcasting refers to postings that transmit information on stances, statements, facts, performances, opinions, and ideas to a dispersed audience. It is about providing information when a response is not expected. Postings are coded as broadcasting instruments if the distribution of information is central. We differentiate if a posting is (rather) broadcasting or (rather) not broadcasting. If it is not evident, the posting is coded as balanced/ambivalent.
Over and above the distribution of information, previous studies have highlighted mobilization as one of the main features of the Web (among the first were Foot and Schneider [36
]). For instance, political parties and candidates use social media to mobilize their supporters to volunteer for their campaigns and also to vote for them (e.g., [37
]). NGOs attempt to mobilize their supporters to donate money (e.g., [40
]), and companies go online to mobilize their stakeholders to buy products, provide feedback on products, and participate in the creation of new products (e.g., [41
When coding whether a posting was perceived as focusing on mobilization, we considered whether it activates, dynamizes, and involves, i.e., whether it “calls for action”. Postings are coded as (rather) mobilizing when more than 50% of the elements in the posting have a mobilizing character and as (rather) not mobilizing when less than 50% of the elements have a mobilizing character. Again, if this is not clearly evident, the posting is coded as balanced/ambivalent.
2.2. Image Management
A social media presence is much about self-presentation and organizations too endeavor to manage their audience’s impression of them (their image and/or brand). Goffman’s (1959, The Presentation of Self in Everyday Life
) work on how individuals attempt to influence the perceptions of other people about them has been influential when understanding political actors’ use of social media platforms [32
]. For instance, Stanyer’s study [43
] (in the field of political communication) shows that politicians use digital media as front regions to present themselves to the electorate. A more recent study by Smith and Sanderson [16
] uses Goffman’s notions of self-presentation and gender display to examine how professional athletes use Instagram. Examining the Instagram feeds of 27 athletes, they found that, similar to print advertisements, women are framed in a subordinate and submissive manner.
Organizations also endeavor to maintain an appropriate impression on social media in order to receive a positive evaluation [44
]. The entire brand communication field and branding per se is ultimately concerned with image management. Moreover, image management is at the heart of visual communication, as images influence how “individuals view (…) products and services” [29
] (p. 122) and they help build a brand’s image, create identification, and link organizations with societal symbols [14
]. Social media are seen as instruments that organizations may use to try to decrease the psychological distance between themselves and their stakeholders [37
Therefore, with this cluster of variables, we analyzed the image organizations convey with their Instagram postings. First, we focused on the much-discussed trend in politics on personalization, which can be connected to the larger trend towards individualization in society [46
]. Is it a personalized image the organization wants to manage or not? The second variable is centered on the debate on the increasing blurring between the private and public sphere [47
]. Does the image the organization wants to manage focus on the private context or not? Third, we analyzed whether celebrities are used to manage the organization’s image. The use of celebrities in advertising is widespread, because this fosters the credibility and attractiveness of the organization, brand, etc., and the celebrity enables the message to stand out from the many other messages [48
]. The question is: Is this also the case online?
To what extent is Instagram used to manage the organization’s professional or personal image? For many years now, there has been a trend towards personalization in mass media and politics [49
], which is also visible online [50
]. Research on political campaigning has shown that parties focus increasingly on the personality of top candidates as the “communication of messages requires a messenger” [51
] (p. 25). This trend decreases the complexity and increases the credibility of a message (see also [49
]). Specifically, it offers a glimpse into the private moments of a politician and can help the voter identify with her/him. This can also be connected to the larger trend of the individualization of society [46
]. We therefore wanted to analyze whether this resonates with the image an organization seeks to manage by means of its postings.
The coding scheme differentiates between postings that are primarily carried by one (or more) single individual(s) (personalized postings) and postings that are primarily carried by many people or those that do not show any people (not personalized postings). Again, if this is not clearly distinguishable, the coding will be reflected as balanced/ambivalent. Furthermore, it is also possible to code whether a leader of the organization, such as the party leader, or the top candidate of a political party, or a CEO of a corporation, is visible or not visible (visibility) in the posting.
Currently, the private sphere has become public, specifically in tandem with the rise of social media platforms [47
]. This is also in line with the affordance of social media in general and Instagram in particular, enabling the easy sharing of the snapshot images of our daily life. The question we thus ask is: Do organizations build on this affordance and on the increasing blurring of the private and public spheres when managing their image on Instagram? This is the rationale behind the variable privatization. We differentiate between a professional context (giving a speech/press conference, shaking hands, being at a rally, etc.) and a private context (family, hobbies, personal matters, etc.).
If the picture focuses predominantly on a person’s professional context, postings are coded as (rather) professional context and (rather) privatized context when a private/personal context dominates. If this is not clearly distinguishable, the posting is coded as balanced/ambivalent and, if no person is visible in the picture, the coding requirements are not met.
With respect to each posting, we recorded whether a famous person is included, namely a V.I.P. (very important person), such as an actress/actor, a musician, or an athlete. This may allow the organization to manage a particular image. The use of celebrities has always been important in marketing [48
]. We also recognized that many people of public interest (celebrities) have profiles on Instagram and organizations use these celebrities to gain users’ attention. An American cigar brand, for example, used the rapper Snoop Dogg as an ambassador for its new brand. Snoop Dogg promoted cigar use extensively on his Instagram account and helped generate awareness of the new cigar [17
If a person of public interest can be identified, the posting is coded as celebrity visible and not visible if this is not the case. If a celebrity is visible in the picture, the name of the identified celebrity is also coded.
The third cluster of variables concerns integration. Instagram is a rather new communication platform for organizations; the question therefore arises of whether Instagram is integrated into an organization’s existing information and communication mix. Previous studies have shown that social media communication is seldom integrated into the overall organizational communication strategy. The use of social media can be described as “mostly experimental and ad hoc” [52
] (p. 303), because resources for social media communications are not available in most organizations. Recent accounts within communication studies have, however, stressed the increasing hybridity [53
] of media systems and highlighted the importance of studying media ecologies [54
] rather than platform by platform. The point is how platforms are integrated into and referenced to each other. Social media platforms especially allow information to be spread and amplified [55
] on many different platforms. We therefore focused on hybridity, shared content, and, in the context of campaigns (in politics, marketing, etc.), if there were references to other types of campaign/communication instruments in the posting.
Social media provide organizations with the opportunity to directly inform and communicate with their stakeholders without having to rely on traditional media outlets [37
]. Chadwick’s 2013 book The Hybrid Media System
stimulated discussions on such hybrid forms of communication. The question thus arises: In what ways are offline and online communication instruments integrated on Instagram? In our study, every posting was analyzed regarding whether a directly recognizable, and therefore explicit, reference to another communication instrument/media was established. We distinguished between traditional (offline) communication instruments/media and new media.
The coding scheme differentiates between an explicit offline media reference in the picture, for example if a newspaper article or a picture of a campaign poster is posted, and an explicit new media reference in the picture, such as hashtags and calls to share pictures. The posting is coded as no explicit reference if neither applies.
● Shared Content
In a study of a Swedish politician campaigning online, Svensson [55
] found that social media are used to amplify messages that emanated on other platforms. Hence, we examined the integration of Instagram into the existing (ecology) of an organization’s information and communication mix. With the variable shared content, we studied whether the shared content was original to the Instagram posting or emanated from elsewhere. We then coded whether the content of the Instagram posting had already been published offline or on the organization’s other social media accounts, such as Facebook, Twitter, and YouTube, or whether it is an original Instagram posting.
When coding for this variable, we distinguished between not shared (original to the Instagram posting) and shared (the postings had already been posted offline or on the organization’s other social media accounts).
● Campaign References
Campaign references refer to all kinds of campaigns, for instance political parties’ regular election campaigns, which are mandatory by law/the political system, or self-controlled marketing campaigns, such as a rollout by a company. The aim of all campaigns is to reach out to an organization’s stakeholders and convince them of the standing of the organization, its activities, products, and services.
Hence, every posting was coded to reflect whether it included a directly recognizable, and therefore explicit, reference to a campaign. For example, the picture features a campaign poster, a voting booth or a hashtag for a new product. We distinguished between explicit campaign references in the posting and no explicit campaign reference.
One of the main attractions of social media platforms is their affordance of interactivity. In this last cluster of variables, we focused on the caption and, especially, on the comments the posting may have generated. Social media have been hailed for empowering citizens and stakeholders, for enabling online public debate, and allowing more participation in decision-making processes [56
]. Riggins [57
] shows that European communicators at large corporations, government bodies, NGOs, and other associations consider stakeholder interactivity and engagement as extremely important: Improvement in an organization’s reputation, the support of long-term relationships, and a better understanding of stakeholder concerns and expectations are highlighted as the greatest benefits. Studies in the field of visual communication have also shown that images can have a rhetorical impact and provide persuasive arguments to their viewers [14
] (p. 122). A study on Instagram by Lee and colleagues [15
] shows the importance of interaction when an organization wants to establish and cultivate organization-public relationships: Social interaction is the primary motive for Korean Instagram users, aged 20 to 39, to use Instagram.
The question that arises is: Is Instagram used to engage in two-way interaction with stakeholders? We studied interactivity in captions and comments. First, we studied the interactive potential of the picture’s/video’s caption content and the comments the posting attracted. We subsequently studied the tonality (whether it was positive or negative), as well as reciprocity, i.e., whether, in the comment field, the users and posters really communicated with each other.
2.4.1. Content of Captions and of Comments
We studied three different aspects of the content of all the captions and comments in order to analyze whether they contribute to an exchange of substantive/essential messages (information). We differentiated between emoticons and captions/comments with (contributing to the exchange of information) or without intrinsic value (such as simple encouragements and applause) [33
In respect of this variable, we therefore coded (a) emoticon(s) only; (b) postings with intrinsic value; and (c) postings without intrinsic value. (a) Emoticon(s) only applies to postings that display a mood/emotion by means of an emoticon; (b) postings with intrinsic value give relevant and substantive information, as they focus on (public/current) issues. They often indicate a statement, an opinion and/or an idea or latest news on upcoming actions or events; and (c) postings without intrinsic value display trivia/nonsense or they only include a simple encouragement, for example, in the case of a politician, a posting such as “go for it” or “you are the best”.
2.4.2. Negative vs. Positive Tonality
Citizens, specifically in their role as voters or consumers, mostly visit an organization’s social media platforms to complain (phenomenon “shitstorm”) or to show their support and good experiences (phenomenon “candystorm”) (cf., [58
]). Here, we distinguished between captions/comments that are predominantly of a negative or a positive tonality. The caption/comment can be text or just an emoticon.
All postings that displayed critique, scandals, conflict, failure, attacks, controversies, pessimistic outlooks, stagnation, or resignation, and negative smileys, etc., were coded as (rather) negative. This is “bad news/information” for the organization. All postings that displayed consensus, the solution of problems, optimistic outlooks, success, approval, and pleasant developments, as well as smileys, etc., were categorized as (rather) positive. If a posting contained no indicators that pointed to a negative tonality, nor ones that pointed to a positive tonality, or they were not clearly distinguishable, the posting was classified as neutral/ambivalent. We did not read sentiments into the postings, but took the written postings at its face value.
Interactivity also means that there is a certain degree of exchange between the participants. Participants need to listen and relate to one another. “Reciprocity can therefore be defined as a basic condition for deliberation. If citizens do not listen to each other and interact (…), there can be no deliberation, only monologue” [59
] (p. 44). In online discussions, this means that the participants respond to one another by commenting on other peoples’ posts and/or comments. Most studies have shown that companies and political actors hesitate to facilitate direct communication with their stakeholders and they hardly ever respond to messages posted on their social media platforms as they fear losing control over the flow of communication (e.g., [3
Here it is coded whether the poster and/or the followers reacted to other peoples’ comments or not, for example by answering questions, giving one’s opinion, highlighting statements of stakeholders, and giving more information on the discussed issue. It is recorded whether the reaction is related to a comment or not related to a comment.
We developed and tested the variables by studying political parties’ use of Instagram during the 2014 election campaign in Sweden, and also in-between elections in 2016 [18
]. In our study, the inter-coder percentage agreement of each of these items fell within the acceptable range, with the vast majority at or above 83% (holsti). Hence, we have tested these variables positively in political settings.
However, as discussed above, we are sure the variables presented can be used in other settings and to analyze the use of Instagram by organizations other than political parties. Further research will help us gain an understanding of the platform and its usage. We also believe that the variables presented can be used as an inspiration to study other social media platforms in today’s fast-changing media and communication landscape. “Older” platforms, such as Facebook, are also changing dramatically in that pictures and videos increasingly dominate. With respect to their stakeholder communication, organizations have to keep in mind that Facebook shows postings with a picture or video to a greater circle of people [11
]. Moreover, the use of Snapchat and Tinder is increasing in the overall Internet population, and specifically among teens [61
]—the future voters—consumers, and supporters. Researchers have to tackle and study “new” platforms that tend to increasingly focus on visual communication. Images are an additional source of information and communication, complementing the written or spoken text, and they help create meaning. They may therefore have the potential to foster online participation and organization-public relationships to a greater extent than rather text-based platforms.
We hope the methodological framework presented will be helpful to the readers of this Special Issue and will promote studies on the use of Instagram and other visual social media from an organizational perspective.