Picture perfect : a mixed-methods analysis of engagement with image-based social media content

With the convenient, constantly updating presence of social media, people are consuming a flood of images daily. Visual material is now available at a user’s fingertips and the nature of this material has evolved; not only featuring models and celebrities, but friends, peers, and ‘influencers’. The detrimental effects of idealised traditional media on wellbeing (particularly body dissatisfaction) are well documented. A growing body of literature has emerged to address social media’s place in this theoretical landscape, with recent research beginning to consider the effects associated with Instagram – an image-based social media platform. Early findings suggest links between Instagram use and body dissatisfaction, however many of the nuances of this platform, including the perception of ‘influencer’ status, have yet to be considered. In this context, the aim of my thesis was to explore some of the complexities associated with the Instagram platform, with a view to defining the mechanisms that potentially underlie the negative effects associated with viewing some Instagram images on factors such as mood, body dissatisfaction and self-esteem. To this end, my thesis firstly comprises a narrative review (Chapter One). Here I, along with my co-authors, trace media effects from traditional to social media; our discussion culminating in an agenda for future research. From this agenda we distilled two core areas for focus in this thesis: firstly, a need for greater understanding of the key image-based behaviours users engage in on social media (namely image-editing, posting, and ‘liking’), with a focus on self-presentation; and secondly, providing an initial examination of the role of ‘influencers’ in the effects of Instagram use informed by our understanding of social comparison theory. The following chapters are presented in two parts, aligning with our two central themes. In Part One (Chapters Two and Three) we sought to investigate the way social media users engage with image-based content. We identified three key behaviours central to interaction with image-based content: editing, posting, and ‘liking’. Our initial focus was broad, either examining social media more generally, or examining the most popular and well-researched platform, Facebook. Taking this approach allowed us to define the aspects of social media use that warranted further investigation in Part Two’s localised examination of Instagram. Part One, Chapter Two presents a study that utilised the Theory of Planned Behaviour model to predict image-editing intentions, controlling for previous editing platform use, Facebook Intensity, and age, and with the inclusion of Narcissism. Analysis of survey data from participants of a variety of ages indicated that editing application use, Facebook Intensity, attitudes, subjective norms, and Narcissism were all significant predictors of intentions to post digitally altered self-images. Attitudes had the strongest association with intentions, while the contribution made by subjective norms likely speaks to the importance of approval and belongingness on social media. Interestingly, perceived behavioural control was not a significant predictor, with our findings suggesting that users experience a great sense of control over whether their posts are edited or not, and that editing-decision making might be unique from other image-based behaviours. The contribution of Narcissism, albeit small, provided support for the prevailing understanding of photo-based activities as self-promoting and superficial. To complete our general investigation of key behaviours we enquired about the factors that adult users of a broad range of ages consider prior to posting or ‘liking’ an image on social media. In Chapter Three we took a qualitative approach, analysing open-ended responses using thematic analysis. Interestingly, egoistic motivations and considerations of the audience were at the forefront of peoples’ minds when posting and ‘liking’ on social media, supporting early suggestions that ‘likes’ may also be a tool for self-presentation. The results culminated in our understanding of posting and ‘liking’ as complex, multi-factored behaviours, which users do not engage in flippantly. This complexity suggests that engaging with images on social media is stressful, particularly when users are invested in the content (i.e., their own image) and anticipating a good response from the audience. Informed by our greater understanding of key image-based behaviours from Part One, in Part Two we aimed to determine the psychological effects of engaging with Instagram images on mood and body dissatisfaction in adult Instagram users. Addressing the possible stress that may be associated with Instagram use (Chapter Three), and the role that investment in one’s content may play, in Chapter Four we developed a set of items to operationalise a new construct: Instagram investment. This construct describes the emotional connection that users have with the content they post and the response they anticipate from the audience. Using these items, we examined the relationships between Instagram investment and a selection of psychological variables, discovering small, though significant, links with negative psychological outcomes including depression and stress. Additionally, Chapter Four examined a mediation model, whereby Instagram investment mediated the relationship between number of followers and low self-esteem. Though small, this effect suggests that popularity is more important to a user’s self-esteem when they are highly invested in their Instagram use. Chapter Five details a follow-up of the Instagram investment construct, exploring its personality predictors in order to identify adult users who may be prone to greater investment in their Instagram content and response. Instagram investment was predicted by higher neuroticism and lower honesty/humility. The link found here between neuroticism and Instagram investment provided support for our initial suggestions that there may be negative implications associated with being highly emotionally invested in your posts to Instagram. Part Two culminates with Chapter Six. Here we detail the first experimental investigation to manipulate ‘likes’ and ‘follow’ statistics to examine the effect of ‘influencer’ status on adult female viewers’ mood, body dissatisfaction, and self-esteem. In doing so, we contribute finer detail to the broader media effects literature. Participants exposed to the same images alongside either ‘influencer’ (high) or ‘peer’ (low) level statistics experienced significantly greater negative mood than those in a nature control group. No differences were found between the influencer and peer groups, suggesting that popularity is not critical to determining image-effects. Instead, it appears that idealised female content has negative implications for the viewer regardless of perceived status. Overall, the findings of my thesis contribute to the broader understanding of media effects by helping to define the mechanisms by which social media influences the user. My studies present the complex underpinnings of image-based behaviours and suggest that these behaviours are much more than simply ‘clicking a button’. The findings help to define the circumstances under which a user can exert control over their online self-presentation and propose Instagram investment as a potential means of detrimental Instagram effects. Our understanding of social comparison theory as it pertains to Instagram is also extended through this research, with perceived target proximity of little importance and the image itself as the key determiner for effects. This thesis sparks broader questions for social media effects and image-based content, as well as conceptualisation of online identity and self-presentation. For example, in the context of selective self-presentation and image-editing, are there ramifications for how users perceive their actual ‘offline’ selves? In the interests of preserving well-being, users may be best served by limiting the amount of idealised content they view on Instagram, however longitudinal methods are the necessary next-step for painting a complete picture of the impact of Instagram on the viewer