What adolescents think of relationship portrayals on social media: a qualitative study

. Background : Understanding the factors in ﬂ uencing adolescents ’ relationship views is important because early romantic relationships often act as precursors for relationships in adulthood. This study sought to examine the types of relationship-focused content adolescents witness on social media and how they perceive its effect on their romantic relationship beliefs. Methods : Sixteen semistructured interviews were conducted with Australian adolescents aged 16 – 19 years who were purposively sampled from a larger longitudinal study. Interview transcripts were analysed qualitatively using constructivist grounded theory. Results : Participants described the types of romantic relationship portrayals they saw on social media, including relationship-focused trends like ‘ Relationship Goals ’ and ‘ Insta-Couples ’ . Participants explained their ability to identify incomplete and unrealistic relationship portrayals, as well as the pressure to share their relationships online in the same incomplete fashion. Views regarding the in ﬂ uence of social media were varied, but most believed social media relationship portrayals had some level of in ﬂ uence on young people ’ s relationship views; some participants believed this occurred regardless of awareness of the incompleteness of the online portrayal. Conclusions : Although participant interview data revealed the pervasiveness of social media relationship portrayals, it also revealed the sophisticated capabilities of adolescents in critiquing online media portrayals.


Introduction
Having grown up in the Internet age, adolescents today live out their lives in an inseparable merger of online and offline worlds. 1 With 92% of those aged 15-16 years and 99% of those aged 16-17 years in Australia having an active social media profile, 2,3 adolescents are now also bringing many traditional developmental tasks, such as social and identity development, into the online environment. 4 Social media has allowed adolescents to engage in selective self-presentation online, posting images and text that reflect their developing identities while gaining feedback and comparing their presentations to those of their peers. 1 Key milestones, such as a young person's first romantic relationship, can now be showcased to their entire social media network with a click of a button. The way social media plays into adolescent romantic relationships is increasingly important given these early relationships are developmentally significant for learning about interpersonal skills, self-identity and sexuality. [5][6][7] Because these early relationships also serve as precursors for relationships in adulthood, 8 as well as being important in their own right for adolescent well-being, 5 it is in public health's interest to explore how adolescents see and interact with relationships on social media in order to understand social media's influence on adolescents' relationship views.
Social media plays an important role in adolescent romantic relationships by facilitating relationship initiation and maintenance. 9 However, beyond this, social media acts as a source of relationship models because users are able to watch their friends' relationships unfold online. Facebook, for example, allows couples to publicly display affection by having a dyadic profile picture, sharing pictures of their activities and by announcing their relationship status to the network by becoming 'Facebook Official'. 10 Particularly for adolescents, sharing this relationship information on social media may be considered customary, because platforms like Facebook ask for relationship status when users create their profiles. 11 Researchers have also posited that these online public displays may be a way that adolescents flaunt their romantic relationships to social networks, in turn boosting selfesteem as they celebrate a merged couple identity online. 12,13 Becoming 'Facebook Official' may, for example, serve as a significant relationship milestone, as users are able to share their joy (and loss) with others, ultimately validating their relationship status by displaying it to their 'networked public'. 14 However, because individuals tend to post especially attractive versions of themselves online, 15 it is possible that those in unhappy relationships continue to post unrealistic couple photographs on social media as a compensatory mechanism or a self-presentation strategy to appear happier in their relationships to others. 12,16 Because social learning theories state that individuals construct their relationship views as a result of observing models or 'scripts' around them, 17,18 it is important to understand what relationship portrayals adolescents witness on social media platforms. Furthermore, we need to know whether adolescents feel influenced by such relationship portrayals online, and whether this influence can construct or alter their views on what a relationship should look like. Having healthy and realistic expectations of relationships is important for adolescents, because unmet expectations, even when knowingly idealised and highly romanticised, can be predictive of poorer relational satisfaction and therefore jeopardise a young person's overall well-being. 19 Although there is evidence indicating the various influences of film and television portrayals of relationships on adolescent relationship views, 20,21 there is little research looking at social media portrayals of relationships. This online context for relationships must be studied given adolescents are exposed to relationships much more on social media than they are offline. 22 This study endeavoured to shed light on these gaps in the literature using descriptive qualitative inquiry; the aim of the study being to explore the views of adolescents towards relationship-focused content on social media and their beliefs about the potential influence of this content on their relationship views.

Methods
This study included 16 adolescents aged 16-19 years who were purposively sampled on the basis of sex, age and experience with social media from a larger mixed-methods longitudinal study on online networks and adolescent relationships conducted in New South Wales, Australia. 23 The University of Sydney granted ethics approval for this study, and all participants provided their own written and verbal consent before the collection of data.
Two authors (MT, LL) conducted 30-to 60-min in-depth semistructured interviews with the participants either by Skype (n = 13) or face to face depending on the participant's preference. Skype interviews are considered to encourage participant openness and allow young people to speak freely, so responses collected over Skype in this study were not expected to differ from face-to-face interview responses. 24,25 The interview guide comprised of openended questions and prompts for the larger study as well as specific prompts designed to explore the objectives of this substudy. For example, some key prompts directed towards participants were 'What kind of relationships do you see online/on social media?' and 'Do you think social media plays a role in how young people form ideas of what relationships 'should be'?' (further prompts are given in Appendix 1). Prompts were developed after a literature review and finalised following pilot testing and analysis of initial emerging themes throughout the duration of the study, identifying the gaps and appropriateness of the initial questions. Participants received an A$20 gift card at the completion of the interview as part of the overall incentive structure of the larger longitudinal study.
Each interview was audio recorded, deidentified and transcribed verbatim, then analysed using a constructivist grounded theory approach in order to create an understanding centred around adolescent perspectives and experiences rather than an existing social theory. 26 During analysis, transcripts were first coded line by line inductively, and then developed into a focused code network using the qualitative software package Dedoose version 7.6.15 (SocioCultural Research Consultants, Manhattan Beach, CA, USA) by two authors (MT, LL). Using a constant comparative method, 27 the codes, concepts and emerging themes were continually compared between and within the transcripts by two authors (MT, LL) and only concluded when there were no major discrepancies between identified themes. The interviews and data analysis were conducted concurrently until data saturation (i.e. the point at which no new themes were observed in the data) was reached. In this study, the point of saturation was reached within the 16 participants, as expected, because a sample size of 12-15 leads to saturation in qualitative data when considering a focused research question. 28 Of the 16 participants sampled for this study, half identified as female (n = 8), with the rest identifying as male (n = 5) or other gender (n = 3). The mean participant age was 18 years, with most of the sample aged 18-19 years (n = 12). Half the participants (n = 8) were in romantic relationships at the time of the interviews. Further demographics and participant IDs are presented in Tables 1 and 2.

Ethics approval
Approval for the study was obtained from the Ethics Committee of The University of Sydney. The procedures used in this study adhere to the tenets of the Declaration of Helsinki.

Results
We identified two main themes: (1) types of relationships adolescents see online; and (2) beliefs about the relationships adolescents see online. Theme 2 consists of three subthemes: (1) incompleteness of online relationships; (2) posting relationships online as the norm; and (3) complexity of influence.

Types of relationships young people see online
Participants described seeing a broad range of romantic relationships online. Typically, popular social media platforms, Facebook and Instagram, were discussed as significant hosts of relationship-focused content. For example, Facebook users have the ability to share their romantic relationships via the 'relationship status' feature or present their relationship through written text posts. However, participants reported that most of the relationship portrayals they saw on social media were predominantly in the form of images (e.g. photographs of a couple in a relationship). These images were often seen on the photo-sharing platform Instagram.
The relationships participants reported seeing online belonged to both people they knew personally (e.g. school friends) and people they did not know offline (e.g. celebrities or friends of friends). In fact, when asked to describe some relationships they saw on social media, most (n = 10) of the adolescents described the relationships of couples they did not know personally. One 19-year-old female participant said: 'I do remember looking at this couple on Instagram just because they are a famous couple. Alexis Ren and Jay Alvarrez?' (Participant 1) Participants explained that many of these relationships were those of 'Insta-couples', couples who had become well known social media celebrities through sharing pictures of their romantic relationship on Instagram. Three female participants named Alexis Ren and Jay Alvarrez as a famous Insta-couple when explaining this concept. These participants explained that Alexis and Jay's relationship had become widely recognised, similar to that of a traditional media celebrity couple in that their relationship was well documented publicly across social media.
Participants also explained that there were specific relationship-focused trends on social media, with participants (n = 9) bringing up the trend of 'relationship goals' during the interviews without being prompted. A specific prompt about 'relationship goals' was developed after several interviews in order to capture information about this specific trend. A 19year-old female participant (Participant 11) explained that 'relationship goals' was a social media trend born out of Internet culture where users share images of seemingly 'idyllic relationships' online, captioning it 'relationship goals'. She explained she would usually see pictures of 'attractive couples living an amazing life', where they would exchange extravagant gifts and travel together. Participants  Beliefs about relationships adolescents see online

Incompleteness of online relationships
All participants raised the issue of incomplete or unrealistic relationship portrayals on social media. Relationships posted on social media were considered to be 'curated' or 'onedimensional' and therefore incomplete, because users choose exactly how to represent their relationship online. Insta-couples in particular were thought to use this to their advantage, portraying their relationship to be 'unbelievably perfect' just to garner fame. One 19-year-old female participant explaining the phenomenon of Insta-couples reported: 'I remember seeing them and being like, 'How is that even real?' A lot of it is probably made up to be a certain way, especially because they became famous through all their couple photos.' (Participant 1) Other participants shared this belief about incompleteness of online relationships, as they explained that social media users only portray the 'best parts' of their relationship online. One 18year-old male participant explained that he and his peers realised this when he saw the discrepancy between a friend's offline and online portrayal of her relationship: 'My friend is in a bad relationship. She will come and tell me this story about what happened today but will make everything look okay on Instagram with all these cute photos. Because of that, we all know that it is not real.' (Participant 16)

Posting relationships online as the norm
Participants commonly referenced the norm of posting images of their relationships on social media. This was explained as an 'expectation' to share their relationship online, as well as more explicit 'peer pressure'. Some participants explained the pressure to post relationships online was a part of the larger pressure they felt on social media, because there was a 'fear of missing a good post'. Other participants raised similar pressures and felt that it was normal not only to showcase relationships online, but also to portray an idealised version of it. They believed young social media users are expected to post images of their interesting and enviable relationships online in a similar fashion to the Insta-couples. One 19-year-old female participant said: 'Instagram is popular for people who are in relationships because they can post couple photos. It's not just Alexis and Jay who post like that. There is peer pressure to be posting happy relationship photos for other people like, '1 year anniversarylook how cute we are'.' (Participant 1) Some participants also believed that there were young people who would engage in relationships just 'for the 'gram'' (a commonly used abbreviation of 'Instagram'). Participants explained that some young people felt pressure to emulate the images of relationships their friends shared online, with some engaging in relationships purely for the purposes of being able to showcase Insta-couple-type images on their own social media profiles. One 17-year-old female participant explained this, stating: 'I know a lot of girls who want to get into relationships for couple photos like that. They don't even care about any of the other stuff. It's all just for social media, social media.' (Participant 7)

Complexity of influence
Most participants (n = 12) believed that social media had at least some level of influence on adolescent relationship views. This belief was common across all background demographic characteristics in this sample, including age, sex or sexuality. It is notable that participants rarely described this influence on themselves and instead spoke about the influence on 'young people' in general. Participants commonly reported 'tweens' and those who 'don't understand real relationships' were the ones most under the influence of online relationships. This was perceived to be because they were unable to identify unrealistic relationships as they lacked experience.
However, four participants went on to explain it was not only those with limited understanding of relationships who could be influenced. They believed social media relationship portrayals could influence their relationship beliefs even when they could identify unrealistic relationships online. Being aware of the unattainable nature of these portrayals is not enough to avoid the influence as one 16-year-old participant, other gender, said: 'It is hard seeing something like that [relationship goals] and actually getting it into your head that your relationship probably won't be as perfect as their relationship seems.' (Participant 2) Some participants explained that social media portrayals may serve as standards against which young people would evaluate their own relationship. Yet, because these standards are unattainable, it would likely lead to problems within their current or future relationships. One 19-year-old male participant said: 'I think seeing 'relationship goals' online takes a toll on actual relationships because people spend so long trying to create an ideal that they know doesn't really exist.' (Participant 10) Some of the participants had ideas about the mechanisms by which social media relationships influenced young people. These participants explained they felt 'bombarded' by portrayals of relationships online, and thus started normalising the unrealistic portrayals. For example, one 16-year-old participant (Participant 2), other gender, believed that these portrayals would gradually permeate into a young person's own relationship views because 'it's always in the back of your head', even when not actively using social media.
Conversely, there were participants (n = 4) who believed relationships on social media were not influential. They explained that trends such as 'relationship goals' were 'not to be taken seriously' and, at most, could inspire potential date or gift ideas, but would not affect any significant part of a relationship. Two participants expressed that they did not see many relationships online in the first place and so felt unaffected by portrayals, both referencing the fact that they do not 'follow' Insta-couples on social media. Other participants explained they viewed these trends as a 'laugh', with one male participant, aged 18 years, explaining the way he used the trend in a humorous way with his girlfriend: 'My girlfriend will tag me in a picture of a family of 10: 2 people with 8 dogs. She will be like, 'relationship goals', because we both absolutely love dogs. That is probably the most I get out of that thing.' (Participant 9) Furthermore, some participants reported that online relationship portrayals could be used in a satirical way. For example, another 18-year-old male participant explained he created a mock 'relationship goals' Instagram account in order to poke fun at the trend. He explained the account was purely for entertainment because the relationship portrayals were so unrealistic that other young people would see the humour in the hyperbolic comparisons: Although the majority of participant responses centred on social media portrayals, some (n = 5) also discussed and compared the portrayals of social media and traditional media portrayals of romantic relationships. They explained that relationship models in film and literature may also be influential for young people, particularly those with limited relationship experiences. For example, one female participant, aged 18 years, explained how traditional media relationship portrayals and norms influenced her behaviour in her early relationships: 'Definitely movies influenced me, because when I first started dating I didn't have much to go by so it was sort of just 'go-with-the-flow'. I think the biggest influence for me would be TV shows, movies and that sort of romanticcomedy style portrayal of relationships.' (Participant 5) Although these traditional media relationship portrayals were considered influential, participants explained that social media portrayals could possibly play into their relationship views to an even greater extent. They explained that the ubiquitous nature of social media, and the ease of sharing a large amount of personal information, would allow other users seemingly more intimate relationship content. One 17-year-old female explained this difference between relationship portrayals on the two media:

Discussion
This study showcases the views of adolescents towards the romantic relationship portrayals they see on social media. Major themes identified were the types of relationships adolescents see online and their beliefs about these relationships, including the incompleteness of online portrayals, the pressure to post relationships online and complexities around the influence of online relationship portrayals. In the interviews, participants used the terms 'online' and 'social media' interchangeably, and so we have also treated the terms as such throughout this section. Our findings expand upon literature in the area, describing another role of social media in adolescent relationships: as a source of relationship models.
An important finding of this study was that participants believed many portrayals of relationships they saw online were incomplete or unrealistic, inferring an ability to critically approach relationship-focused content online. This challenges studies that suggest young people are unlikely to know the difference between realistic and unrealistic media content. 29 Our findings are supported by other studies that indicate adolescents use a range of critical skills and perspectives when interpreting media content, skills that develop with age and experience. 20,30 In addition, because young social media users construct their online identity themselves, they are aware of the curated nature of the content they see on social media. 31 Our findings also indicate these critical skills may have developed through observing online-offline discrepancies between relationships around them. This reasoning capacity suggests that not all adolescents should be automatically considered naïve online media consumers. However, it is important to note that older adolescents, like the participants in the present study, may have more sophisticated reasoning and critical skills than younger adolescents. 29 This study also reveals a different way to approach the issue of adolescents and social media. As noted in the findings, social media use by young people is not simply a matter of 'exposure' or 'consumption': young people themselves produce and circulate social media content within the context of friendships and relationships. Viral Internet trends such as 'relationship goals' and the existence of Insta-couples rely on engagement by social media users to proliferate; otherwise they would not be trends in the first place. Young social media users encourage the proliferation of this content, potentially due to the norms mentioned in our findings, or simply because they genuinely enjoy sharing such images. Some participants who engaged with the 'relationship goals' trend did so in the context of parody and playfulness or to communicate intimacy, as in the case of Participant 9 who would use it to joke with his girlfriend about their love for dogs. This engagement with relationship-focused content online may, in fact, aid young people's understanding of romantic relationships, rather than simply projecting a one-dimensional model they copy. Therefore, online portrayals should not always be considered as intrinsically harmful, or otherwise problematic. Furthermore, a critique of 'unrealistic' relationship representation may, in some cases, be a critique of racialised or heteronormative social media tropes, as noted in the example of Participant 16, where two non-heterosexual friends set up a satirical 'relationship goals' Instagram account in which they mimicked a heterosexual Insta-couple.
The role of these social media models on participant relationship views is complex, with participants having mixed views on its influence. Most believed social media was influential to some extent at least, attributing this influence to the constant and abundant portrayals of relationships online, further amplified by the accessibility of the Internet through mobile devices. These findings align with our current understandings of social media influence on young people, 20,32,33 and suggest that scripts witnessed online are incorporated into adolescent relationship views. However, there were several participants who felt social media did not influence them at all. This difference may be explained by social desirability bias, because adolescents may underreport the influence of social media portrayals, especially 'unrealistic' ones, due to perceived judgement. 34 This may also explain why our participants would immediately explain the influence of social media on other adolescents and not on themselves, a phenomenon common in social media studies with young people. 35 However, it is also likely that social media, like all other influences, affects adolescents to different degrees.
As evident in the findings, social media cannot be considered a completely independent influence on adolescent relationship views. The relationship portrayals that young social media users are exposed to online are likely projecting relationship norms that already exist in the offline world. For example, adolescents witness the relationships of their offline friends when online. Because adolescent engagement with social media is tied to their broader peer groups, the norms that are reinforced online do not deviate much from peer norms offline. 31 Similarly, the other type of relationships adolescents encounter on social media, those belonging to people they do not know personally (e.g. Insta-couples), are akin to the norms that already exist in traditional print, broadcast or entertainment media. The participants' descriptions of relationship portrayals of Instacouples, for example, resembled that of traditional media's celebrities: readily displayed across popular media and seemingly perfect. Consequently, we emphasise that relationship-focused content on social media such as Instacouples and 'relationship goals' should not be considered as isolated influences. Many of the norms projected by these portrayals have already existed offline for years and may not be inherently negative for adolescent well-being, but should be considered as one facet of the broader context in which contemporary young people learn about relationships. Young people's attitudes to relationships are shaped by multiple and complex sources of relationship models. Understanding the role of social media, just one source of external influence among others (traditional media, parents and friends), can aid our understanding of the way young people approach romantic relationships overall.

Limitations
Like all opt-in studies, only participants who were interested in the study and felt confident enough to share their thoughts would be likely to participate. This would have excluded non-English speakers and adolescents who did not feel comfortable talking about romantic and sexual relationships. Although this study did not identify any differences in participant responses based on demographic background, future studies with larger samples should be able to ascertain whether the findings reflect the experiences of a broader population sample of Australian adolescents, and indeed adolescents from other countries and cultures. In addition, only half the participants self-identified as being 'in a relationship' at the time of the research. Therefore, their perceptions of 'unrealistic' social media representations of relationships were not drawn from lived experience, but likely from a comparison to their observations of other people's reports.

Implications
A major strength of this qualitative study was its ability to present the nuances of social media influence as perceived by adolescents themselves. These findings allow for a more informed understanding of adolescent romantic relationship views, important for those who support the healthy development of young people, including parents, clinicians, educators and policy makers. In particular, the complexity of these social media relationship portrayals uncovered by this study highlights the need for adults working with young people to be conscious of diverse representations of sexuality and relationships in social media, including satirical and/or humorous content. These findings could inform how social media can be used for education and health promotion programs, encouraging healthy relationship views among young people and, in turn, healthy relationships. For example, relationship education programs could counter the incomplete relationship portrayals study participants described seeing on social media by producing similar content that instead portrays relationships that model a range of diverse 'relationship goals', including mutual respect, consent and personal autonomy.

Conflicts of interest
The authors declare that they have no conflicts of interest.