“He has a way gayer Facebook than I do”: Investigating sexual identity disclosure and context collapse on a social networking site

Abstract

This study extends previous research into social networking sites (SNSs) as environments that often reduce spatial, temporal, and social boundaries, which can result in collapsed contexts for social situations. Context collapse was investigated through interviews and Facebook walkthroughs with 27 LGBTQ young people in the United Kingdom. Since diverse sexualities are often stigmatized, participants’ sexual identity disclosure decisions were shaped by both the social conditions of their online networks and the technological architecture of SNSs. Context collapse was experienced as an event through which individuals intentionally redefined their sexual identity across audiences or managed unintentional disclosure. To prevent unintentional context collapse, participants frequently reinstated contexts through tailored performances and audience separation. These findings provide insight into stigmatized identity performances in networked publics while situating context collapse within a broader understanding of impression management, which paves the way for future research exploring the identity implications of everyday SNS use.

Keywords

coming out context collapse identity impression management LGBTQ self-presentation sexuality social media social networking sites stigma

First, consider a scenario where a member of an LGBTQ¹ group marches in an annual pride parade and, while shouting cheers about marriage equality, is seen by co-workers to whom he has not yet disclosed his homosexuality. Now imagine a pansexual² woman, who has only disclosed her orientation to a few friends, making a post on Facebook about same-sex marriage to which much attention is drawn by her uncle, who responds with a series of negative comments. In both instances, unintended audiences received these expressions of identity. Goffman (1959) noted the ways in which individuals tailor their behavior for certain audiences within specific contexts, thereby often avoiding in-person situations like the one described above. However, with certain qualities of social networking sites (SNSs) augmenting the reach and salience of identity expressions, the possibility of such occurrences online is heightened. This phenomenon has been identified as context collapse,³ a flattening of the spatial, temporal, and social boundaries that otherwise separate audiences on SNSs (boyd, 2011).

Drawing on Goffman’s (1968) approach of examining groups for whom impression management is intensified due to their position in society, this article accomplishes two main goals toward attaining a better understanding of context collapse. First, exploration of LGBTQ young people’s experiences relating to sexual identity disclosure on Facebook reveals the way that stigmatized identity performances can be adapted to SNSs’ technological affordances to harness, manage, or prevent context collapse. SNS features combine with stigmatizing social conditions so that experiences of homophobia or fears of being discredited shape online self-presentation decisions. LGBTQ young people’s tactics for rebuilding contexts illustrate the way they deal with these challenges while opening up possibilities for better understanding other stigmatized populations’ experiences of context collapse.

Second, this study aims to extend conceptualizations of context collapse by positioning it within the broader symbolic interactionist framework of impression management (Goffman, 1959). Similar to the collapse of a physical structure, the deterioration of social context has environmental influences, receives reactions from the individuals involved, and may alter future behavior in efforts to prevent another collapse. By examining this larger picture of social exchanges, the experience of context collapse can be understood as an event through which individuals might intentionally or unintentionally have their identity redefined across audiences. Available strategies for the prevention of involuntary context collapse illustrate both the limitations and opportunities available to users for adapting to or modifying the technological architecture of SNSs in order to reinstate contexts for identity performances.

Theoretical roots of context collapse

Goffman (1959) described impression management as the process through which social actors tailor performances for “front stage” situations containing an audience and relax behavior in “backstage” regions where this audience is absent. Performances consist of intentional expressions, involuntary expressions given off, and background information forming one’s personal front. Through this combination, an actor attempts to define the situation in such a way as to make claims about being a certain type of person, providing the audience with expectations about the actor’s behavior and the responses it warrants. Since expressions that form the definition of the situation are adjusted to a specific setting, time, and audience, an outsider entering the front stage region mid-performance threatens to invoke a sort of context collapse. This is because the actor’s present performance may contradict the definition of the situation previously established with this new audience member, thereby discrediting the actor’s prior identity claims. As such, context collapse can be understood as an event, or episodic occurrence, within a specific situation where certain aspects of the setting and identity performance influence its likelihood since “some contexts are more porous than others” (Davis and Jurgenson, 2014: 477).

Focusing on face-to-face interactions, Goffman’s concept of impression management does not address how new technologies affect social interactions. As online platforms consisting of personal profiles, publicly articulated connections, and multiple modes of interacting with user-generated content (Ellison and boyd, 2013), SNSs bring audiences together as users build vast online networks. These collections of people connected through technology can form networked publics with affordances that enhance the persistence, replicability, scalability, and searchability of actors’ performances (boyd, 2011). These affordances lead to the dynamic of collapsed contexts where “the lack of spatial, social, and temporal boundaries makes it difficult to maintain distinct social contexts” (boyd, 2011: 49). The technological architecture of SNSs can dissolve the boundaries of front stage and backstage regions, increasing the porousness of contexts. This creates “environments in which contexts are regularly colliding” (boyd, 2011: 50) when multiple audiences receive the same identity performance.

As noted by Papacharissi (2009), “While the architecture of social networking sites is suggestive, it does not have to be inherently limiting” (p. 203). Despite experiencing a setting more conducive to context collapse than others, users adapt their behavior to the architecture of SNSs to reap the benefits of participation, such as social capital (Vitak, 2012; Vitak and Ellison, 2012). On Twitter, the entirety of one’s audience can only be imagined (Marwick and boyd, 2011), which may lead some users to include only enough personal information to portray the degree of authenticity necessary for maintaining a network of followers. One example of this is the playful use of hashtags to express emotion without appearing too personal (Papacharissi, 2012). By treating the entire SNS as front stage, these users are able to maintain a single definition of the situation through consistent self-expression.

Strategies for avoiding context collapse are complicated by SNS features that offer varying degrees of access to identity performances. On Facebook, users can tailor their privacy settings so that only individuals they have intentionally added to their network (“friends”) can view their profile information and online activity. With its real name policy (Facebook, 2014), users generally add contacts whom they have already met offline (Baym, 2010) or, if adding new acquaintances, those who are friends of friends (Dutton and Blank, 2013). This results in massive networks of people with whom a certain definition of the situation has already been established. Despite this personalized audience, Facebook’s networked structure and algorithms limit users’ control over the timing and display of identity performances (Hogan, 2010; Zhao et al., 2013). This increases individuals’ likelihood of having audience members witness identity expressions that they would not otherwise encounter in environments with less permeable contexts.

Tactics for preventing context collapse on SNSs with personal and known audiences vary in the limits they place on individual expression. Adolescents give targeted performances through social steganography, encoding publicly visible messages so they are only understood by certain audiences (boyd and Marwick, 2011; Oolo and Siibak, 2013). Similarly, many college students’ Facebook photos only provide enough visual context to be meaningful to those present when they were taken (Mendelson and Papacharissi, 2011). Applying stricter measures, adult SNS users often refuse co-workers’ friend requests or bar access to personal identity performances by creating separate professional accounts (Vitak et al., 2012). When users have accumulated many overlapping audiences and feel increasingly subject to social surveillance, they place greater limitations on self-expression (Brandtzæg et al., 2010). This can result in users displaying only the most benign identity performances acceptable to the lowest common denominator of their network (Hogan, 2010). While these findings speak to the prevalence of context collapse, they also demonstrate the possibility of reinstating contexts. To extend this research, it is necessary to further investigate how SNSs’ technological architecture can combine with the social conditions shaping identity performances and audience receptions, which may increase the intensity of context collapse experiences and one’s motivation to avoid them.

Managing stigmatized identities and privacy

Stigma can be understood as an attribute that establishes difference from others by creating a discrepancy between normative expectations about individuals and their actual identity (Goffman, 1968). In relation to LGBTQ people, Foucault (1979) and others (Jagose, 1996; McIntosh, 1968; Weeks, 1996) have identified discourses in Western society that led to the stigmatization of certain sexual acts whereby they came to be understood as reflections of identity, creating the homosexual person and other “deviant” roles. Consequently, the stigmatization of individuals with diverse sexual identities has often led them to be viewed as immoral, diseased, and unlawful people, since they do not meet normative expectations relating to gender, sex, and sexuality (Butler, 1990).

Disclosure of one’s stigma poses challenges for impression management as it alters the definition of the situation and carries uncertainty regarding others’ reactions (Goffman, 1968). For LGBTQ people, disclosure often takes the form of coming out of the closet (Plummer, 1996; Sedgwick, 1990), an ongoing process of revealing one’s sexual orientation, which may never be fully complete. Goffman (1968) noted that many stigmatized individuals live between the two extremes of zero and full disclosure, presenting multiple selves to different audiences. Similarly, LGBTQ people frequently apply an approach of strategic outness (Orne, 2011), first assessing a specific social situation before determining whether to disclose.

As LGBTQ people make disclosure decisions, they often consider these stigmatizing social conditions and align identity performances to projected outcomes, but such predictions become complicated by the affordances of SNSs. Even if individuals treat entire SNSs as front stage, upholding a single definition of the situation by abstaining from sexual identity expressions, online contacts can reflect a user’s identity (Jernigan and Mistree, 2009; Walther et al., 2008) and interactions can accumulate as expressions given off (Goffman, 1959). Remaining closeted can also preclude individuals from encountering allies and other LGBTQ people online, connections of which hold benefits for identity formation and reinforcement (Gray, 2009; Laukkanen, 2007). Therefore, some individuals may choose to come out to certain contacts by displaying stigma symbols (Goffman, 1968), indications of identity intended to be recognized only by friendly audiences. With these varying levels of disclosure, LGBTQ Facebook users must constantly manage the flow of information and renegotiate identity expressions as new contacts are added to their network (Cooper and Dzara, 2010).

Grave instances of online discrimination (e.g. Huffpost Gay Voices, 2011; Parker, 2012) intensify the need for LGBTQ individuals to prevent context collapse in the form of unintended sexual identity disclosure. Since young people who are outed on SNSs are often the target of discriminatory remarks and threats of physical violence (Varjas et al., 2013), regulating personal information about sexual identity is not only a matter of impression management but also of safety and privacy. Nissenbaum (2009) views privacy as the fulfillment of informational norms: expectations about what will be disclosed to whom, which vary between contexts. However, Facebook’s frequently changing and often confusing privacy settings defy informational norms as users regularly share more than they intend (Acquisti and Gross, 2006; Liu et al., 2011; Madejski et al., 2011). Therefore, it is possible to see how a greater understanding of context collapse and strategies for its prevention can preserve users’ privacy. Skilled Facebook users are less threatened when others affect their self-presentation since they feel better equipped to address these discrepancies in their identity performance (Litt et al., 2014). As users develop larger and more diverse Facebook networks, they also become more likely to use Friend Lists to separate contacts (Vitak, 2012). Some individuals even circumvent Facebook’s terms and conditions, and its intended open design, by employing aliases and conducting regular “wall cleanings” to remove the accumulation of personal information (Raynes-Goldie, 2010). These actions restore a degree of privacy by counteracting features of SNSs that increase the likelihood of context collapse (e.g. persistence, searchability).

Aside from Lim et al.’s (2012) study of juvenile delinquents who negotiate self-presentation to gang members, family, and the authorities simultaneously on Facebook, little existing research addresses how stigmatized identities requiring increased vigilance are managed in relation to context collapse. This article examines the impression management of individuals with stigmatized identities on Facebook. It tracks LGBTQ young people’s decisions regarding sexual identity disclosure, their experiences of context collapse, and their preventative strategies to avoid unintended future occurrences of context collapse.

Methods

Participants were recruited through LGBTQ student groups at 11 universities in the United Kingdom, which were chosen for their diversity with regard to student demographics, location, heritage, and programs. This yielded a sample of 27 individuals, including 12 international students, from a variety of backgrounds representing different ethnicities, socioeconomic statuses, and life histories. Participants’ ages ranged from 18 to 25 with an average age of 20. There were 15 who were identified as female, 11 as male, and 1 as agender.

The study focused on Facebook as the most popular SNS in the United Kingdom at the time (Blank and Groselj, 2012; Wilson et al., 2012) and because its “friending” process creates a knowable collection of audiences. Many early adults have matured throughout Facebook’s decade-long existence, and this was reflected in participants’ long-term use of the platform, averaging 5 years with all having accounts for at least 3 years. Early adults were chosen for their higher likelihood of having already self-identified their orientation in mid-to-late adolescence (Grov et al., 2006; Savin-Williams and Diamond, 2000), positioning them within the process of further establishing their desired level of disclosure. While some participants were hesitant to label their orientation, all were accustomed to using a term to explain their sexuality to others: 14 identified as gay, 5 as bisexual, 4 as lesbian, 2 as queer, 1 as pansexual, and 1 as asexual. Although a limitation of this sample is the missing participation of trans individuals, the inclusion of diverse sexualities and gender expressions added perspectives outside of homonormative (and cisnormative) gay and lesbian culture (Duggan, 2002). While all participants were out to at least one close friend, they varied in their level of disclosure online. In all, 12 included intentional indicators of sexual identity on Facebook and were out to everyone in their friend network, 9 were out to some and not others, and 6 chose not to include any indications of sexual identity on Facebook even if they were out to others in offline situations.

Semi-structured in-person interviews averaging 60 minutes were conducted to obtain thick descriptions (Geertz, 1973) about identity disclosure and context collapse. Interviews combined a biographical approach (Nilsen, 2008), asking participants to discuss personal experiences of sexual identity disclosure, with practical queries about Facebook use and online self-presentation. Discussions were facilitated by having participants navigate through their Facebook accounts while responding to questions. Similar to photo elicitation (Harper, 2002), this provided a memory aid and an extra layer of visual data as Facebook walkthroughs were video recorded using Apple’s QuickTime Player. Looking at accounts in conjunction with participants allowed for deeper discussion about disclosure decisions and provided context for exchanges visible on their Facebook profiles. Interviews were coded and analyzed through a pragmatic application of grounded theory methods, which referenced existing literature while allowing for themes to emerge from data (Bryant, 2009; Strauss and Corbin, 1998). Examining individual codes alongside biographical stories ensured that participants’ narratives of navigating SNSs were understood within the overarching social settings of their lives (Riessman, 1993), including the social conditions influencing disclosure of stigmatized sexual identities.

Findings and discussion

The interview data provided many insights as to how LGBTQ individuals negotiate sexual identity disclosure on Facebook. While some participants discussed taking advantage of the platform’s affordances to come out across their network, others spoke about involuntary disclosure as identity expressions became visible to unintended audiences. As discussed below, both scenarios can be understood as instances of context collapse with the latter leading participants to adopt strategies for preventing similar occurrences in the future. Prevention strategies mainly fell into two categories: tailoring identity performances or separating audiences, both of which were implemented to rebuild contexts.

Identity disclosures and experiences of context collapse

Context collusions

To examine decisions that LGBTQ young people make regarding sexual identity disclosure, participants were asked to identify indications of sexual identity that were present on their Facebook accounts. When asked if he had voluntarily shared anything, Marco⁴ responded, “I can show you.” He played a YouTube video about an 8th grader named Jonah who had been bullied about his sexuality and was contemplating suicide⁵. Translating from Italian, Marco read the message he had posted for all his Facebook contacts, “I’m gay, like Jonah, and even though—fortunately—I haven’t lived the same situation in that way, I just wanted to come out. And Jonah, you’re not alone.”

Despite Facebook’s promotional “Stories” page featuring Kai Bailey’s (2012) experience of using the platform as a tool for “coming out in a single click,” only three participants had announced their sexual identity through a highly visible post using the Timeline feature. In contrast, 10 participants had completed the “interested in” profile field to indicate same-sex attraction and many perceived it to be a subtler expression even though profile information remains prominently displayed over time while Timeline posts slip from view relatively quickly. Some participants who refrained from providing this information criticized Facebook’s binary options of being interested in men or women as too limited to fit their sexual identity while others felt that explicitly stating sexual preferences should be reserved for dating sites. However, participants also noted that an incomplete “interested in” field could indicate that a user was not heterosexual. A solution for coupled individuals was to instead disclose their sexual identity by completing the “relationship status” field. Henrik shared that doing so helped him come out to acquaintances from his hometown, “… If you are in a relationship it’s less, you are like less vulnerable because you are clearly gay and being successful at it.”

These overt and intentional expressions of sexual identity can lead to redefinition of the situation throughout one’s entire social network on Facebook. Henrik explained that this was his intention, “If I meet people, I want them to as quickly as possible know that I’m gay without having to talk about it because I feel that when people know I’m gay, it’s over with.” By expressing his sexual identity on his profile, Henrik added to his personal front (Goffman, 1959) and was able to bring this information into future social situations. His boyfriend confirmed that such online displays affected subsequent in-person interactions:

I don’t want to run the risk of having a confrontation in real life so if I have it on Facebook, they can take it in, deal with it themselves, and then it’s over and done with and I never had to say anything.

Both participants’ emphasis on disclosure being over speaks to how broadcasting sexual identity accelerates the coming out process, which can be facilitated by collapsed contexts. Davis and Jurgenson (2014) separate context collapse into two types of events: context collusions, through which actors intentionally flatten contexts, and context collisions where contexts unintentionally “come crashing into each other” (p. 480). Voluntarily coming out on Facebook can be understood as context collusion and, although context collapse is a neutral occurrence, this increased ability to spread identity disclosures may have positive outcomes. It may allow individuals to quickly overcome phases of guilt, secrecy, and solitude that can be experienced when closeted (Plummer, 1996) and may also add knowledge to discourses in ways that rearrange power structures (Foucault, 1979) so that information about sexual identity can no longer be used to subvert or silence individuals. However, the actual outcomes of voluntarily coming out on Facebook may not be this straightforward. Many individuals disclose their stigma to ease social situations but subsequently feel the need to engage in acts of covering to keep this quality from “looming large” (Goffman, 1968: 125). Even after coming out to everyone on his Facebook, Matt deleted pictures of him kissing a man to avoid negative reactions: “I think people were like, ‘Yeah, I don’t care if you’re gay, don’t like shove it down my throat’.” Similarly, 10 participants stressed that they did not want to flaunt their sexual identity or have it be the first detail others noticed on their profile.

Context collisions

While the salience of sexual identity may be modifiable, coming out across one’s network precludes the option of strategic outness (Orne, 2011) as a way of completely keeping this information from sensitive or homophobic audiences. For this reason, many participants chose to display expressions of sexual identity only to some contacts or none at all. However, even those who took great measures to rid Facebook of information about their sexual identity found that indications still crept into view in the form of group memberships, page “likes,” events, photos, and friends’ posts. These involuntary expressions given off (Goffman, 1959) pose the threat of context collapse, or more specifically context collision (Davis and Jurgenson, 2014), as they may be viewed by audiences whose definition of the situation involves a different understanding of participants’ sexual identity. Maria, who is not yet out to her extended family in South America, experienced this when she scrambled to remove a friend’s comment that said “Oh them gays” on a photo of her. Despite warning her friends not to post anything on Facebook, an indication of her sexual identity eventually surfaced.

Participants’ experiences of identity expression and its outcomes showed context collapse to be an event rather than an end state. It is a transitory step toward either redefinition of the situation or a myriad of corrective, face-saving activities allowing for the initial definition to be upheld (Goffman, 1959, 1972). With his 2000 Facebook friends including many religious family members and acquaintances from a Jewish camp, Nate engaged in “lowest common denominator culture” (Hogan, 2010: 9) to keep his Facebook free of information about his gay identity. He maintained scarce details on his profile and posted only uncontroversial quotes from poems. However, his volunteer work for LGBTQ groups began to appear as updates about his event attendance were broadcast through Facebook’s newsfeed feature. With these increasing occurrences of context collapse, Nate felt he would rather risk redefinition of his sexual identity than expend further energy attempting to control this information: “And then I was like, who fucking cares? To give that much power to other people and a program seems crazy.” Having recently made this decision, Nate was still waiting for his contacts to take notice of these automatic posts while not intentionally add any “overtly gay” indicators: “I’m not there yet,” he explained.

Other participants scrambled to save face through techniques of information control (Goffman, 1968) that included deleting, hiding, and untagging posts. Since Elizabeth spent years constructing a highly visible, gay identity on her Facebook, she described actively deleting content posted by friends alluding to events that could lead others to perceive her as bisexual: “[Having] waved the big lesbian flag, I was really embarrassed when I started to slip up and accidentally have one night stands with men, and I definitely remember editing what was visible on my profile page.” Despite research and theories indicating that fluidity and instability of sexual identity may be a common and widespread experience (Rust, 1993; Seidman, 1994), Elizabeth felt that explaining her bisexual activity would be “a sort of second coming out.” Although she eventually spoke to her friends about it in person, she noted, “I suppressed [it] on Facebook before I was ready to talk about that level of complication.” Her experience describes not only the challenges involved in redefining a firmly established definition of the situation but also Facebook’s limitations for expressing identities that are fluid or less recognized by mainstream understandings of sexuality.

While participants were frequently able to manage context collapse retroactively, sudden or unintentional redefinition of the situation threatens to discredit individuals as having misrepresented themselves (Goffman, 1959). Since this can disrupt social interactions, threaten relationships, and may even endanger individuals, it is often desirable to avoid context collisions. This was the case for many participants who were not ready to disclose their sexual identity to those who would view it in a stigmatizing manner. They employed preventative behaviors prior to and alongside sexual identity expressions, which were often successful in maintaining differential definitions of the situation across multiple audiences.

Strategies for preventing context collapse

Tailoring performances

One set of tactics for context collapse prevention involved tailoring identity expressions so they would be received differently among audiences. Participants often achieved this by maintaining the ambiguity of potential sexual identity indicators through humor, such as changing one’s relationship status to being “married” to a best friend, or by posting messages that heterosexual allies or groups with certain political beliefs also shared. In spring 2013, a sea of red flooded Facebook as an advocacy organization, the Human Rights Campaign, encouraged users to change their profile picture to its logo in support of marriage equality as the US Supreme Court deliberated same-sex marriage laws (Kleinman, 2013). Ana, an American student who had not disclosed her bisexual identity to all of her contacts, was not anxious about joining this movement. She explained, “It doesn’t say, ‘I am gay’. It says, ‘I support gay rights’.”

Since these indications of sexual identity may be too subtle to reap the benefits of connecting with other LGBTQ people and friendly audiences, several participants’ accounts also included posts rife with stigma symbols (Goffman, 1968). William, who was not out to his parents or many people in his hometown, welcomed a friend’s post on his Timeline that used Photoshop to depict William marrying the male lead singer of a boy band. Although this portrayal was laden with symbols stereotypically associated with homosexuality (e.g. Lady Gaga, a celebrity associated with LGBTQ activism, as the marriage officiant), William was confident that his homophobic audiences would not see this as an indication of his gay identity: “A lot of my family members who wouldn’t approve are the kind of people who don’t really recognize homosexuality as something that exists.” Similarly, Talan noted that his family members in Africa do not assume he is gay when they see Facebook photos of him in drag. He shared, “My friend was showing [my aunt] all the pictures of me dressed in really skimpy fancy dress and she was like, ‘Wow, I have no idea why you’re dressed like that’.” Although they could not know for certain that such messages left in plain sight would only resonate with specific audiences, these participants engaged in a sort of social steganography (boyd and Marwick, 2011), relying on stigma symbols to indicate sexual identity to some contacts and not others.

While the success of participants’ encoded messages may have been due to cultural differences as well as variations in media exposure, age, and location, it is also likely that their audiences played a role in upholding the established definition of the situation. Not only did Goffman (1959) note that audience cooperation with an actor’s performance saves time and emotional energy, he also identified that individuals allot to each other a certain level of civil inattention—taking notice of others in public but not scrutinizing them (Goffman, 1966). Echoing boyd and Marwick’s (2011) finding that teenagers expect others to view their profiles for social purposes and not with the intention of surveillance, participants asserted that close examination of profiles (“Facebook stalking”) was a violation of privacy norms. With her sexual identity undisclosed to her extended family, Erin admitted, “If someone trawled through my Facebook they probably could, like, get a picture that I was gay but you’d have to be really persistent. They’d have to be a real stalker so I’m not too concerned about it.” Participants noted that the only exception was if they thought a Facebook acquaintance might also identify as LGBTQ, then it was permissible to investigate further for the purpose of connecting as friends or romantically. This was how Henrik first noticed his future boyfriend: “He has a way gayer Facebook than I do.” Overall, participants who used ambiguous and encoded messages in their identity expression assumed both that indicators would remain unrecognizable to unintended audiences and that these audiences would afford them civil inattention by continuing to reinforce existing definitions of the situation.

Separating audiences

The other set of tactics used to prevent context collapse involved reconstructing social contexts by separating audiences for identity performances. This approach took advantage of different affordances of SNSs to establish boundaries of varying fortitude depending on the sensitivity of the information being shared and the particular audience. Generally, participants began by tailoring their privacy settings and “friending” practices within Facebook, which have been described in prior research respectively as targeted disclosure and network regulation (Vitak and Kim, 2014). In all, 13 participants used Friend Lists to limit certain contacts’ access to content, including the “interested in” profile field and information about their relationships. Five made efforts to only “friend” individuals within their generation while avoiding older contacts who might be in a position of authority or have less tolerant views. Kyle regularly updated a tiered system of lists that only allowed his closest friends to see all of his content. He explained,

If I stop talking to someone, they’ll slowly go down in privacy settings and then eventually, if I think that there’s no point in me having them on Facebook, they go. And then new people I meet will get their privacy settings upgraded.

However, many participants were frustrated by Facebook’s labyrinth of privacy settings, experienced pressure to remain friends with certain contacts out of professional or familial obligation, or found that “weeding out” friends required a great deal of effort.

For these reasons, several participants separated audiences across different SNSs. Six turned to Twitter as a site where they broadcasted LGBTQ content and subscribed to LGBTQ-themed accounts under their real names. Individuals benefited not only from having different audiences on Twitter, with only a few contacts from Facebook, but also from the particular context provided by the platform’s social and technological affordances. Although participants admitted that audiences to whom they had not disclosed their sexual identity could easily find their Twitter account, they felt that expressions on Twitter reflected their identity less than on Facebook. Emily described, “You haven’t got a whole profile, it’s more like purely for thought.” With the platform’s large volume of content, participants stated that LGBTQ-related messages also seemed subtler on Twitter and that it was a more appropriate space than Facebook for this subject matter. Holly, who does not express her lesbian identity on Facebook but frequently tweets about LGBTQ topics, affirmed that Twitter “is more political generally as a medium… If someone saw something on here then they’d be more likely to think that it was just a political statement rather than actually sort of who you are.” For these participants, Twitter upholds informational norms (Nissenbaum, 2009) better than Facebook because they have a clearer consensus regarding the type of content that is acceptable for sharing and how it might be interpreted in this context. This, along with minimal personal information on profiles, provides a space that may reduce occurrences of context collapse.

Participants requiring thick boundaries around particularly personal or less accepted identity expressions turned to more anonymous online spaces. As mentioned in the introduction, Jessica has disclosed that she identifies as pansexual to only some of her close friends. She spoke about a Tumblr account that she keeps secret from everyone she knows offline:

My Tumblr’s like a retreat almost, it’s kind of like—sometimes I think it’s who I want to be on Facebook … It’s not a secret life because I know that I can be quite vocal about [my sexuality] on Facebook but in terms of actually expressing it and like, looking at images of things that to me are erotic, it’s almost like this is where my sexuality is—where no one can find it …

As a tool for curating and displaying content, Tumblr allows users to connect with others and shape their online self-presentation through the practices of blogging and reblogging (Leavitt, 2013). Unlike SNSs with real name policies, it affords users the ability to obscure identifying features through the use of pseudonyms, which continue to be essential for the creation of context-specific identities online (Hogan, 2013).

Similarly, Mackenzie has come out to few people offline but frequently uses Tumblr and discussion boards to express an asexual identity. This combination resonates with findings that such online spaces can support community building and identity affirmation for LGBTQ people where there is no offline equivalent (Dehaan et al., 2013; Gray, 2009). Comparable to Elizabeth’s hesitation to express a bisexual identity on Facebook, Mackenzie asserts that the site is not very useful for disclosing an identity outside both heterosexual and homosexual mainstreams:

There’s not much opportunity to tell people on Facebook that you are asexual … If you were gay […] you could say “I am interested in men” if you were a man and that would be fairly easy … If you were going to come out to people as asexual there would always be, you know, you would always have to tell them overtly.

For this reason, Mackenzie maintains Facebook and Twitter accounts for interacting with general audiences, which are kept completely separate from activity in more anonymous online spaces. Altogether, participants’ use of multiple SNSs illustrates their construction of backstage contexts, which serve specific purposes or allow for different types of identity expression, while maintaining Facebook as front stage to their entire network.

Conclusion

These LGBTQ young people’s experiences have illustrated how performances of stigmatized identities unfold on an SNS where technological affordances augment the propensity for contexts to collapse. Through participants’ accounts, it has become evident that their sexual identity disclosure decisions were shaped by both the social conditions in which their sexuality would be interpreted (e.g. homophobic family members) and the technological architecture of SNSs used for identity expressions (e.g. Facebook in contrast to Twitter). By closely applying Goffman’s impression management lens to these experiences, it is possible to understand context collapse as a recurring event that (a) can be harnessed to voluntarily redefine identity across multiple audiences; (b) can be experienced unintentionally, leading either to involuntary redefinition or actions to reinstate the pre-established definition of the situation; and (c) may be prevented through tactics of tailoring identity performances or separating audiences to rebuild contexts for interactions.

By demonstrating approaches for addressing context collapse and creating positive interactions on SNSs, these findings reinforce research regarding the benefits of these sites for young people (boyd, 2014; Ito et al., 2010). Such findings could have practical application in helping to shape media literacy curricula by teaching users about context collapse risks, uses, and prevention strategies. This could also be tailored to LGBTQ organizations to support individuals in the process of coming out, as these young people have demonstrated that it increasingly involves the use of online platforms. Participants’ struggles to protect their privacy emphasize a need for SNS designs that incorporate user-friendly privacy settings, and this study’s findings may contribute to policy discussions regarding privacy on SNSs. At minimum, it is in Facebook’s best interest to improve its privacy features before users further disperse their activity across platforms.

The study is limited in its focus on university students, who may not be representative of the rest of the population, and its voluntary recruitment, which presents the potential for self-selection bias. These and other limitations call for further research to determine the theoretical transferability (Lincoln and Guba, 1985) of concepts discussed in this article. Given worldwide differences in political and social attitudes toward diverse sexualities and the rapid evolution of communication technologies, this study only addresses a small portion of the vast need for more research about LGBTQ individuals’ experiences in networked publics. Additionally, SNS users may be required to make identity disclosure decisions relating to other qualities that are often stigmatized (e.g. health issues, employment status) or may embody intersecting identities (e.g. religious and sexual identities, such as LGBTQ Christians) that can intensify or complicate identity expression. Therefore, this study’s findings pave the way for future research involving LGBTQ people and other populations to further explore the impression management and identity disclosure implications of everyday SNS use.

Footnotes

Acknowledgements

As this research was conducted at the Oxford Internet Institute, University of Oxford, I would like to thank Dr. Rebecca Eynon and the rest of the OII’s faculty for their guidance. I also wish to thank my colleagues, friends, faculty at QUT, and the anonymous reviewers from New Media & Society for their excellent insights and suggestions. Deepest thanks to the study’s participants for sharing about their experiences.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

Author biography

Stefanie Duguay is a doctoral researcher in Digital Media Studies and a member of the Social Media Research Group at the Queensland University of Technology. A recent graduate of the Oxford Internet Institute, her research focuses on the formation of publics and counterpublics through social media with particular attention to its implications for sexual identity disclosure and queer visibility.

References

Acquisti

Gross

(2006) Imagined communities: awareness, information sharing, and privacy on the Facebook. Privacy Enhancing Technologies 4258: 36–58.

Bailey

(2012) Essay: coming out in a single click. Facebook Stories. Available at: http://www.facebookstories.com/stories/3790/essay-coming-out-in-a-single-click (accessed 22 February 2013).

Barker

Richards

Bowes-Catton

(2009) “All the world is queer save thee and ME …:” defining queer and bi at a critical sexology seminar. Journal of Bisexuality 9(3–4): 363–379.

Baym

(2010) Personal Connections in the Digital Age. Cambridge: Polity Press.

Blank

Groselj

(2012) The nature of social network sites: adoption, use and types. In: American sociological association annual meeting, Denver, CO, 17–20 August.

boyd

(2011) Social network sites as networked publics: affordances, dynamics, and implications. In: Papacharissi

(ed.) A Networked Self: Identity, Community, and Culture on Social Network Sites. New York and London: Routledge, pp. 39–58.

boyd

(2014) It’s Complicated: The Social Lives of Networked Teens. New Haven, CT and London: Yale University Press.

boyd

Marwick

(2011) Social steganography: privacy in networked publics. In: 61st annual ICA conference, Boston, MA, 26–30 May. Available at: http://www.danah.org/papers/2011/Steganography-ICAVersion.pdf (accessed 5 December 2012).

Brandtzæg

Lüders

Skjetne

(2010) Too many Facebook “friends”? Content sharing and sociability versus the need for privacy in social network sites. International Journal of Human-Computer Interaction 26(11–12): 1006–1030.

10.

Bryant

(2009) Grounded theory and pragmatism: the curious case of Anselm Strauss. Forum: Qualitative Social Research 10(3). Available at: http://www.qualitative-research.net/index.php/fqs/article/view/1358/2850 (accessed 12 May 2013).

11.

Butler

(1990) Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge.

12.

Cooper

Dzara

(2010) The Facebook revolution: LGBT identity and activism. In: Pullen

Cooper

(eds) LGBT Identity and Online New Media. New York: Routledge, pp. 100–112.

13.

Davis

Jurgenson

(2014) Context collapse: theorizing context collusions and collisions. Information, Communication & Society 17(4): 476–485.

14.

Dehaan

Kuper

Magee

(2013) The interplay between online and offline explorations of identity, relationships, and sex: a mixed-methods study with LGBT youth. Journal of Sex Research 50(5): 421–434.

15.

Duggan

(2002) The new homonormativity: the sexual politics of neoliberalism. In: Castronovo

Nelson

(eds) Materializing Democracy: Toward a Revitalized Culture Politics. Durham, NC: Duke University Press, pp. 175–194.

16.

Dutton

Blank

(with Groselj

) (2013) Cultures of the Internet: The Internet in Britain (Oxford Internet Survey 2013 Report). Oxford: University of Oxford.

17.

Elizabeth

(2013) Challenging the binary: sexual identity that is not duality. Journal of Bisexuality 13(3): 329–337.

18.

Ellison

boyd

(2013) Sociality through social network sites. In: Dutton

(ed.) The Oxford Handbook of Internet Studies. Oxford: Oxford University Press, pp. 151–172.

19.

Facebook (2014) Facebook’s name policy. Available at: https://www.facebook.com/help/292517374180078 (accessed 6 March 2014).

20.

Foucault

(1979) The History of Sexuality: An Introduction, vol. 1. London: Allen Lane.

21.

Geertz

(1973) Thick description: toward an interpretive theory of culture. In: Geertz

(ed.) The Interpretation of Cultures. New York: Basic Books, pp. 3–30.

22.

Goffman

(1959) The Presentation of Self in Everyday Life. Garden City: Doubleday.

23.

Goffman

(1966) Behavior in Public Places: Notes on the Social Organization of Gatherings. New York: Simon & Schuster.

24.

Goffman

(1968) Stigma: Notes on the Management of a Spoiled Identity. Englewood Cliffs, NJ: Prentice Hall.

25.

Goffman

(1972) Interaction Ritual: Essays on Face-to-Face Behaviour. Harmondsworth: Penguin Books.

26.

Gray

(2009) Out in the Country: Youth, Media, and Queer Visibility in Rural America. New York and London: New York University Press.

27.

Grov

Bimbi

Nanin

(2006) Race, ethnicity, gender, and generational factors associated with the coming-out process among gay, lesbian, and bisexual individuals. Journal of Sex Research 43(2): 115–121.

28.

Harper

(2002) Talking about pictures: a case for photo elicitation. Visual Studies 17(1): 13–26.

29.

Hogan

(2010) The presentation of self in the age of social media: distinguishing performances and exhibitions online. Bulletin of Science, Technology & Society 30(6): 377–386.

30.

Hogan

(2013) Pseudonyms and the rise of the real-name web. In: Hartley

Burgess

Bruns

(eds) A Companion to New Media Dynamics. Malden, MA: Wiley-Blackwell, pp. 290–308.

31.

Huffpost Gay Voices (2011) Jamie Hubley, gay 15-year-old Ottawa, Canada teen commits suicide, cites depression, school troubles. Huffington Post, 17 October. Available at: http://www.huffingtonpost.com/2011/10/17/jamie-hubley-commits-suicide_n_1015646.html (accessed 7 March 2013).

32.

Ito

Baumer

Bittanti

(2010) Hanging Out, Messing Around, and Geeking Out. Cambridge, MA: MIT Press.

33.

Jagose

(1996) Queer Theory: An Introduction. New York: New York University Press.

34.

Jernigan

Mistree

(2009) Gaydar: Facebook friendships expose sexual orientation. First Monday 14(10). Available at: http://firstmonday.org/ojs/index.php/fm/article/view/2611/2302 (accessed 6 May 2014).

35.

Kleinman

(2013) How the red equal sign took over Facebook, according to Facebook’s own data. Huffington Post, 29 March. Available at: http://www.huffingtonpost.com/2013/03/29/red-equal-sign-facebook_n_2980489.html (accessed 15 May 2013).

36.

Laukkanen

(2007) Young queers online: the limits and possibilities of non-heterosexual self-representation in online conversations. In: O’Riordan

Phillips

(eds) Queer Online: Media, Technology & Sexuality. New York: Peter Lang, pp. 81–100.

37.

Leavitt

(2013) Reblog if: information resharing on a massive creative social media platform. In: 63rd annual conference of the International Communication Association, London, 17–21 June.

38.

Lim

Vadrevu

Chan

(2012) Facework on Facebook: the online publicness of juvenile delinquents and youths-at-risk. Journal of Broadcasting & Electronic Media 56(3): 346–361.

39.

Lincoln

Guba

(1985) Naturalistic Enquiry. Beverly Hills, CA: SAGE.

40.

Litt

Spottswood

Birnholtz

(2014) Awkward encounters of an “other” kind: collective self-presentation and face threat on Facebook. In: Proceedings of the 17th ACM conference on computer supported cooperative work & social computing, Baltimore, MD, 15–19 February, pp. 449–460. New York: ACM.

41.

Liu

Krishnamurthy

Gummadi

(2011) Analyzing Facebook privacy settings: user expectations vs. reality. In: Proceedings of the 2011 ACM SIGCOMM conference on Internet measurement, New York, 2–4 November, pp. 61–70. New York: ACM.

42.

McIntosh

(1968) The homosexual role. Social Problems 16(2): 182–192.

43.

Madejski

Bellovin

Johnson

(2011) The failure of online social network privacy settings. Columbia University Computer Science Technical Reports. Available at: http://academiccommons.columbia.edu/catalog/ac:135406 (accessed 15 February 2013).

44.

Marwick

boyd

(2011) I tweet honestly, I tweet passionately: twitter users, context collapse, and the imagined audience. New Media & Society 13(1): 114–133.

45.

Mendelson

Papacharissi

(2011) Look at us: collective narcissism in college student Facebook photo galleries. In: Papacharissi

(ed.) A Networked Self: Identity, Community, and Culture on Social Network Sites. New York: Routledge, pp. 251–273.

46.

Nilsen

(2008) From questions of methods to epistemological issues: the case of biographical research. In: Alasuutari

Bickman

Brannen

(eds) The Sage Handbook of Social Research Methods. London: SAGE, pp. 81–94.

47.

Nissenbaum

(2009) Privacy in Context: Technology, Policy, and the Integrity of Social Life. Stanford, CA: Stanford University Press.

48.

Oolo

Siibak

(2013) Performing for one’s imagined audience: social steganography and other privacy strategies of Estonian teens on networked publics. Cyberpsychology: Journal of Psychosocial Research on Cyberspace 7(1). Available at: http://www.cyberpsychology.eu/view.php?cisloclanku=2013011501&article=7 (accessed 5 July 2013).

49.

Orne

(2011) “You will always have to ‘out’ yourself:” reconsidering coming out through strategic outness. Sexualities 14(6): 681–703.

50.

Papacharissi

(2009) The virtual geographies of social networks: a comparative analysis of Facebook, LinkedIn and ASmallWorld. New Media & Society 11(1–2): 199–220.

51.

Papacharissi

(2012) Without you, I’m nothing: performances of the self on Twitter. International Journal of Communication 6: 1989–2006.

52.

Parker

(2012) The story of a suicide. The New Yorker, 6 February. Available at: http://www.newyorker.com/reporting/2012/02/06/120206fa_fact_parker?currentPage=all (accessed 7 March 2013).

53.

Plummer

(1996) Symbolic interactionism and the forms of homosexuality. In: Seidman

(ed.) Queer Theory/Sociology. Cambridge, MA: Blackwell, pp. 64–82.

54.

Raynes-Goldie

(2010) Aliases, creeping, and wall cleaning: understanding privacy in the age of Facebook. First Monday 15(1–4). Available at: http://firstmonday.org/ojs/index.php/fm/article/view/2775/2432 (accessed 15 February 2013).

55.

Riessman

(1993) Narrative Analysis. Newbury Park, CA: SAGE.

56.

Rust

(1993) “Coming out” in the age of social constructionism: sexual identity formation among lesbian and bisexual women. Gender & Society 7(1): 50–77.

57.

Savin-Williams

Diamond

(2000) Sexual identity trajectories among sexual-minority youths: gender comparisons. Archives of Sexual Behavior 29(6): 607–627.

58.

Sedgwick

(1990) Epistemology of the Closet. Berkeley, CA and Los Angeles, CA: University of California Press.

59.

Seidman

(1994) Queer-ing sociology, sociologizing queer theory: an introduction. Sociological Theory 12(2): 166–177.

60.

Strauss

Corbin

(1998) Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. Thousand Oaks, CA: SAGE.

61.

Varjas

Meyers

Kiperman

(2013) Technology hurts? Lesbian, gay, and bisexual youth perspectives of technology and cyberbullying. Journal of School Violence 12(1): 27–44.

62.

Vitak

(2012) The impact of context collapse and privacy on social network site disclosures. Journal of Broadcasting & Electronic Media 56(4): 451–470.

63.

Vitak

Ellison

(2012) ‘There’s a network out there you might as well tap’: exploring the benefits of and barriers to exchanging informational and support-based resources on Facebook. New Media & Society 15(2): 243–259.

64.

Vitak

Kim

(2014) “You can’t block people offline:” examining how Facebook’s affordances shape the disclosure process. In: Proceedings of the 17th ACM conference on computer supported cooperative work & social computing, Baltimore, MD, 15–19 February, pp. 461–474. New York: ACM.

65.

Vitak

Lampe

Gray

(2012) “Why won’t you be my Facebook friend?”: strategies for managing context collapse in the workplace. In: Proceedings of the 2012 iConference ACM, Toronto, ON, Canada, 7–10 February, pp. 555–557. New York: ACM.

66.

Walther

Van Der Heide

Kim

S-Y

(2008) The role of friends’ appearance and behavior on evaluations of individuals on Facebook: are we known by the company we keep? Human Communication Research 34(1): 28–49.

67.

Weeks

(1996) The construction of homosexuality. In: Seidman

(ed.) Queer Theory/Sociology. Cambridge, MA: Blackwell, pp. 41–63.

68.

Wilson

Gosling

Graham

(2012) A review of Facebook research in the social sciences. Perspectives on Psychological Science 7(3): 203–220.

69.

Zhao

Salehi

Naranjit

(2013) The many faces of Facebook: experiencing social media as performance, exhibition, and personal archive. In: Proceedings of the SIGCHI conference on human factors in computing systems, CHI’13, Paris, 27 April–2 May.