The two-step cluster analysis procedure resulted in a four cluster solution. For a visual depiction of the outness clusters, see Figure 2.
Cluster 1—High Overall Outness. Cluster 2—Low Overall Outness. Cluster 3—Less Out to Family. Cluster 4—More Out to Family. We examined cluster differences in Facebook networks with respect to both network composition and network connectivity. The results are reported in Table 1 and Figure 3. Table 1 depicts the proportion of network alters in each relational subgroup across the four cluster types.
Figure 3 depicts the proportion of alters in each relational subgroup illustrated in node size as well as the proportion of edges between each pair of relational subgroups illustrated in edge size. With respect to network composition, we found that young people in Clusters 2 Low Overall Outness and 3 Less Out to Family had the smallest percentage of family members represented in their Facebook networks 7. Young people in Cluster 1 High Overall Outness had the lowest percentage of church alters represented in their Facebook networks 0.
Network composition and connectivity by cluster type. Note: Node size is relative to the proportion of alters in each relational subgroup. Edge size is relative to the proportion of edges between each pair of relational subgroups. Note: The percentages reflect the aggregated number of alters in the Facebook networks of participants in that cluster type. With respect to network connectivity, Cluster 2 Low Overall Outness showed the most distinct patterns.
While the other cluster types showed high connectivity between school and family and between family and LGBT community, Cluster 2 showed lower connectivity between these network subgroups, but higher connectivity between school and neighborhood. Surprisingly, Cluster 3 showed strong connectivity between LGBT community and family, even though individuals in this cluster type were less out to family than to others in their lives. Finally, Cluster 1 High Overall Outness showed the most evenly distributed connections between the network subgroups, as illustrated in its visual appearance as the most complete diamond.
These quotes are listed separately by cluster type below. Embracing and accepting that part of me so openly sometimes upsets them. So this account… is the only place I can truly express myself freely. This account has received some negative feedback from more conservative family and friends. I then added them to the limited visibility privacy group I created to keep people who feel this way from seeing my raunchier posts. Ideally, I think it is smart to separate professional colleagues, family, and fellow alumni from my LGBT circle and those I am sexually intimate with.
One Facebook account is to keep in touch with all my associates, friends, past coworkers and teachers. One Facebook is to study astrology. And to trick my mom and aunts into thinking they have a peek into my actual life. So I now have a very limited second personal profile with only a few photos, most of which include myself with professionals in my field or doing volunteer service. Just the mask I allow employers and my employment network to see I posted a picture of my partner and I and received a lot of positive comments and likes. People are very supportive of our relationship and enjoy looking at our pictures.
Real Girls: Shifting Perceptions on Identity, Relationships, and the Media
I currently have females tracking my Facebook page talking down to about me to other friends, family turning on me for something that was out of my hands. My relationship was almost jeopardized. As these quotes illustrate, participants reported a range of ways they use Facebook as well as positive and negative experiences online. Some participants used multiple accounts to manage the information that was shared with diverse relational subgroups, particularly information that related to their LGBTQ identities, was sexual in nature, or otherwise reflected behavior they did not want their entire Facebook networks to know about for example, alcohol or drug use.
Work and family were the two contexts participants mentioned most frequently with respect to the need for multiple accounts. Aside from the use of multiple accounts, other participants talked about monitoring or censoring what they post about online out of concern over how others would respond. Participants also described a range of experiences with respect to information they shared on Facebook. However, participants more commonly identified negative responses, such as transphobic or homophobic comments. These responses illustrate how information shared across diverse relational subgroups can lead to interpersonal conflict and stress for LGBTQ young people.
Contesting the Norm? Live-in Relationships in Indian Media Discourses
Overall, participants from all cluster types described these identity management strategies and stressors related to context collapse. Study findings highlight the complex relationships between outness, Facebook network structure, and online experiences for LGBTQ young people. Using cluster analysis, we were able to identify four distinct patterns in how LGBTQ young people manage outness to different relational subgroups.
The results showed that most participants reported uniformly high or low outness to all relational contexts, but other participants reported differential levels of outness to family as compared to other people in their lives. Nearly two-thirds The high proportion of participants in this cluster type should be understood within the context of the study sample, which was comprised of young adults in an urban setting who had identified as LGBTQ for some time.
Given that young people in rural areas experience more negative reactions to coming out [ 14 ], the percentage of young people in this cluster type would likely be lower in a rural sample. Outness to family emerged as an important variable that distinguished between cluster types, which is consistent with previous research documenting the importance of family as a relational context within which the presence or absence of acceptance and support shapes outness [ 17 , 18 ] and mental health [ 19 , 20 ] for LGBTQ young people.
In particular, youth in the Less Out to Family cluster type appear to negotiate outness to relational subgroups in complex ways. These negotiations can be understood as a resilience strategy for minimizing experiences of stigma and discrimination in an effort to preserve psychological well-being [ 52 ]. At the same time, if these disclosure negotiations reflect active identity concealment in order to avoid experiences of family conflict or rejection, they are likely to negatively impact psychological well-being [ 12 ].
We did not directly examine the level of acceptance and support participants experienced in different relational contexts, and this is an important direction for future research.
- Abducting Casey (Romantic Suspense)!
- 1. Activia: It Starts Inside!
- Continental Reminiscences;
- 10 Ad Campaigns Inspired By Powerful Consumer Insights.
- Cultivation theory - Wikipedia?
Additionally, previous research [ 19 ] has found that family support increases over time for LGBTQ young people who lack this support early in adolescence. Thus, it is possible that a greater number of participants would have fallen into the Less Out to Family cluster type in a younger sample.
Interestingly, some young people in the current study reported being more out to family than others in their lives. Participants in this cluster type had the lowest proportion of LGBTQ alters in their networks, which supports this interpretation.
Real Girls : Shifting Perceptions on Identity, Relationships, and the Media
This cluster type represents an interesting deviation from the types of negotiations related to outness heretofore examined in research, which typically focus on LGBTQ youth who censor information about their identities from non-affirming families [ 17 , 18 , 26 , 27 ]. The use of NameGenWeb enabled us to capture and analyze Facebook network data in a fraction of the time and with significantly less respondent burden than these processes would have taken using traditional methods. Given that participants were young adults, it is perhaps not surprising that school represented such a large proportion of network alters.
Representation of LGBT network alters varied by cluster type: those with uniform levels of high or low outness had the highest proportion, while those who were more out to family had the lowest proportion. This finding belies the stereotype that LGBTQ young people who are closeted are more likely to be isolated from LGBT community and highlights how identity disclosure and relational contexts may be associated in counterintuitive ways [ 52 ]. With respect to Facebook network connectivity, visualizations illustrated a number of similarities and differences between cluster types.
Consistent with research finding that people with more interconnected relational subgroups tend to feel more connected to their network as a whole [ 32 ], we found that LGBTQ young people with uniformly high levels of outness showed the most connectivity between different relational subgroups. Participants in these clusters likely experienced less need to keep their different relational contexts separate, given that their identity disclosure was uniformly high across these different subgroups. By contrast, network connectivity was less evenly distributed between relational subgroups for participants in the other three cluster types, for whom disclosure may be a more complicated process.
For participants in three of the four cluster types High Overall Outness, Less Out to Family, and More Out to Family , network connectivity was greatest between school and family alters, followed by connectivity between family and LGBT community alters. Although this pattern was consistent with overall higher connectivity for the High Overall Outness cluster type, it was striking and counterintuitive for the Less Out to Family cluster type.
Participants in the High Overall Outness and More Out to Family clusters seem to have little reason to keep their LGBT community and family relational contexts separate, as they likely disclose their identities within each of these groups. However, participants in the Less Out to Family cluster type should be motivated to keep their LGBT community and family contexts separate given that these connections increase the risk of identity disclosure to family.
Further, given that Facebook connections typically represent offline interactions [ 4 ], this finding suggests that these participants are negotiating the overlap of these two relational contexts in their offline lives. This raises a number of questions, such as in what settings this overlap takes place and whether or not these families are aware of the LGBT identities of these network alters particularly given that these connections likely give them access to the Facebook personae of LGBT alters. Overall, this finding highlights that outness to and contact between relational subgroups may operate in different and counterintuitive ways.
The most visually distinct cluster type was the Low Overall Outness cluster, which showed dramatically lower connectivity between family and other relational contexts as well as dramatically higher connections between school and neighborhood. With respect to family, young people in this cluster type appear to show a more expected relationship between outness and network structure, such that family members are excluded from Facebook networks altogether or are isolated from other relational subgroups when they are included. The connectivity between school and neighborhood may indicate that these young people primarily attended school close to where they lived, such as cases where they attended community college or where the highest level of educational attainment was high school or below.
Given research on the negative consequences of outness for sexual minority men of lower socioeconomic status [ 16 ], it is possible that participants in this cluster represent more economically marginalized LGBTQ young people relative to the other cluster types. These patterns may reflect the importance of LGBT relational contexts for these participants, for whom connection to LGBT family and community may be an important compensatory source of support if their families are not supportive of their identities.
Although rich in their ability to communicate information about the Facebook relational contexts of LGBTQ young people, these network data do not tell us as much about how participants experience and manage their identities on Facebook. The illustrative quotes identified in the qualitative data provide additional context in this respect. For example, a participant in the High Overall Outness cluster identified that she still felt the need to create a separate Facebook page for family, illustrating how identity management may still be complicated for youth who are highly out but whose relational contexts are not uniformly affirming or accepting.
Further, young people in this cluster who disclose their identities openly on Facebook may experience negative responses to this disclosure, as in the case of the participant who experienced a negative reaction to her post about a Disney princess. Across all cluster types, participants reported use of identity management strategies including creating multiple accounts and monitoring their online self-presentation and identified emotional and relational costs associated with negative responses to online expression of their identities when they did choose to share this information with Facebook network members.
These patterns illustrate the complex interpersonal dynamics experienced by LGBTQ young people in social media environments characterized by context collapse. This study had several important limitations. First, the sample was comprised of young adults who had identified as LGBTQ for at least several years and who lived in a major urban area with multiple opportunities to access LGBTQ resources and community. Study findings would likely be quite different for LGBTQ people of different ages, at different points in the identity development process, and in different regions of the country.
Second, given that most of the youth in this study fell in the High Overall Outness cluster type, analysis for the other three cluster types was based on relatively small sample sizes; thus, study findings should be viewed as preliminary and exploratory. Third, our approach to examining the Facebook network structure focused on the cluster level by aggregating individual networks within each cluster type. Given that there was variability in the sizes of the networks in our sample, this means that larger networks were, in a sense, over-represented in the cluster-level patterns.
The results may have differed if we had used an individual level approach to comparing network structure. Fourth, Facebook friends accumulate over time, which shapes the opportunity structure for connections to form between diverse relational subgroups; for example, if you move across the country, there is little opportunity for your hometown friends and family to meet your current friends or LGBT community.
As other researchers have identified for example, Reference [ 32 ] , this means that some of the assumptions typically made in social network analysis may not hold true for Facebook networks. We did not ask participants in this study about their geographic histories or other factors that could shape this opportunity structure, and this is an important future direction for online social network research. The key quotes provide some insight into these experiences, but more in-depth examinations of these experiences are warranted. Future research could also explore the potential impacts of network connectivity between diverse relational contexts and alters with conflicting political or religious attitudes, such as whether this connectivity has the potential to lead to attitudinal change or increased polarization.
Finally, we were specifically interested in how LGBTQ young people manage their identities on Facebook, but young people are increasingly using a range of other social media platforms, such as Instagram, Snapchat, Grindr, and Tumblr. Ironically, this participant was from the Low Overall Outness cluster type—the type that showed the most distinct patterns with respect to Facebook network composition and connectivity, and for whom the segmentation between family and other relational contexts appears to be the strongest.