Data Excess in Digital Media Research

What do we do with the excess data from our research? ‘Excess’ – particularly in digital media research – is inevitable. It emerges in the research process as the ‘debris’ and ‘leftovers’ from planning, fieldwork and writing; the words cut from drafts and copied to untouched and forgotten files; and the data archived but never analysed or published. From our conversations with colleagues, to our call for contributors, we repeatedly heard researchers’ stories of digital data overflow, as they shared a collective sense of excess data as something more than that which is simply left out of formal research outputs. Digital excess, in particular, holds discursive flexibility: it points to abundance and possibility but also to our failure to control or contain information. Excess data matter, but how and why they do is somewhat opaque and largely underexplored.

This book, Data Excess in Digital Media Research, is a dedicated collection that pays attention to excess data. We position ‘excess’ as a conceptual, methodological, ethical and pragmatic challenge and opportunity for digital media research – we examine what happens when media researchers return to their surplus archives and explore the labour and affects surrounding data overflow and excess. We suggest that data excess is – or should be – a central concern for digital media scholars because of the methodological characteristics of digital media research, the ‘research ethos’ around data excess and the unexpected affects and ‘hauntings’ of excess data. This introduction provides an overview of these concerns and outlines each chapter.

There is a tendency to conceive of unpublished research material as ‘excess’ data, as research resources that are surplus to requirements. In this chapter, I challenge this view by rethinking and critically reframing how we might make sense of research data that, for whatever reason, does not find its way into public presentation or publication. The position I take in this chapter is that we might conceive of unpublished data as operating in dynamic relation to what is or might be presented or published from it, thus serving as vital contextual curtilage shaping ‘field formation’ (how one understands both the immediate context of a research project and the wider context in which the research project is situated). I develop this argument in three steps. First, I look at grounded theory approaches and ask: how do we amass qualitative research data, and how do we determine how much data is enough data? Second, I then turn to consider (less commonly considered) questions: what are the processes for converting this data into publication, and how do we conceive of data that remains unpublished? Third, I then revisit a large collaborative research project of my own that gathered more data than was ultimately reported on, using this as an opportunity for renewed critical reflection on data sets and the productive possibilities of ‘unused’ research data. In considering these possibilities, I draw from the philosophical ideas of Jacques Derrida and the methodological reflections of science and technology studies scholar John Law.

In this chapter, we explore ‘unanticipated excess’ through the lens of our own doctoral research projects, which are presented as distinct vignettes: Reade’s digital ethnography of young women’s relations with ‘fitspo’ (fitness inspiration) content on Instagram, Moran’s social media ethnography of African young people in Australia and Lyall’s show-and-tell interviews with users of digital self-tracking devices. While our projects differ in many ways, we share research practices that did not fully anticipate the challenges of digitalised research fields. In coming to terms with our unanticipated excess, we reflect on inescapable moments and uneasy feelings from our fieldwork. In so doing, we argue that excess need not be considered a ‘failure’ – to establish boundaries, to filter data or to engage in objective analysis – but should rather be seen as an important part of reflexive research practice. Excess holds possibilities and potentials to foster care and camaraderie between digital scholars and can push us and our work – empirically, methodologically and ethically – in new directions. It also presents an opportunity to continue to champion integrity over production as we move forward in our personal and collective research journeys.

Digital ethnographers acknowledge that online spaces are always co-produced within the social, political, material and sensory – never distinct from what we may think of as ‘offline’. However, in documenting our fieldwork (e.g. fieldnotes, screenshots and recordings) and representing our findings in research outputs, scholars tend to draw more firm boundaries around our object of study. The excess, the digital life on the margins of digital ethnography often entangled with the fieldwork site, is cut away to present a neatened case study that can be analysed. In this chapter, I examine the excess and ‘unrelated’ screenshots I took during a digital ethnography project in 2020 to explore what these ‘offcuts’ can offer in contextualising my encounters with the short-form video app TikTok. Over nine months in 2020, I observed healthcare workers using the app to share health information and analyse their content. At the same time, with the pandemic unfolding across the world, I was scrolling through the news on Twitter, watching press conferences from health authorities, sharing funny TikToks with friends and receiving information in a family group chat. This layering of everyday experiences of the pandemic forms part of how I sensed and experienced TikTok content during my digital ethnography. I examine these ‘excess’ screenshots to think through the always more-than-digital boundaries of digital ethnographic fieldwork. I reflect on the messy entanglement of digital ethnography, where my own digital practices – intensified by COVID-19 lockdown conditions – and the broader conditions they emerged from, became inevitably enmeshed with my research practice.

When I think about the relative abundance of data in digital media research, I often recall a remark I overheard – it’s just like digital hoarding? – while presenting digital ethnographic research at a youth mental health conference. I study youth mental health and social media and often interact with scholars who privilege more systematic, empiricist ways of understanding young people’s lives, in contrast to my own iterative approaches to collecting, archiving and creating with what feels like endless digital and non-digital data. In this chapter, I return to the remark and explore digital hoarding as a productive and generative concept for exploring excess digital data in digital ethnography research and fieldwork. I argue for interpreting excess data as inevitable in digital fieldwork and something that is potentially, but not necessarily always, generative for digital media researchers. By sharing research on how people talk about hoarding possessions and reviewing hoarding in digital contexts, I discuss how digital media research may mirror, or at least attempt to mirror, the complex yet everyday quotidian digital practices, including digital hoarding, of people and the digital cultures they create and participate in. This chapter considers the implications of digital hoarding in fieldwork and how embracing digital excess and hoarding may disrupt systematic approaches to digital organisation or attempts to predetermine hierarchical relationships between data. In doing so, I do not attempt to formalise my methods as a digital ethnographer but instead aim to focus on creativity, curiosity and fluidity as guiding principles in digital fieldwork.

This chapter assesses the possibilities of integrating computational network analysis into ethnographic fieldwork. Grounded in examples from an ethnographic case of travel vlogging in Singapore, I discuss how participant observation in urban areas can be combined with network cluster analysis. The huge variety of data collection tools available to present-day media researchers has prompted a plethora of disparate data points and a multitude of ethical dilemmas. Computational network analysis provides manifold avenues for combining interpretations of localised meaning-making with visual evidence about networks of recommended videos on the popular video-sharing platform YouTube. The dissemination of video content is largely shaped by the platform’s recommender system. Exploring how the YouTube recommender algorithm drives the circulation of vlogs about Singapore’s touristic Clark Quay area, this chapter brings to light that the act of recommending a video is based on semantic similarities among video titles and video keywords. Furthermore, the methodological reflections presented in this chapter demonstrate that the combination of computational network analysis and ethnographic fieldwork provides holistic understandings of how highly mediated tourist places are unbound from their physical settings and drawn into platform ecologies, consisting of local areas of production, algorithmic technologies, disseminated place images and platform audiences. As algorithmic mediation plays an important part in accessing platform content about politics, health, culture and entertainment, a myriad of everyday practices is affected by recommender systems. Such algorithmic technologies raise multiple ethical concerns about the accountability of platform owners for a fair and balanced distribution of content on the internet.

This chapter considers how phenomenology and ethnography can be combined as an interpretive strategy for studying mobile media practices in everyday life. We argue that haptic methodologies in mobile media research generate an overflow of data that are often difficult to capture in written form. The dilemma of data excess is discussed in the context of a 4-year research project on gendered mobile media practices, exploring how mobile devices are often intertwined with the experience of bodily safety in urban environments. Our primary aim was to investigate the relationship between networked connectivity, mobile media and perceptions of risk in terms of bodily experience, including affective or emotional feelings. This chapter focuses on the haptic methods and techniques used during home visits with participants and the various ways that the volume and density of data are subsequently thinned out to become manageable and publishable. We first situate our research within haptic and mobile media studies, followed by an explanation of phenomenology and how it can be used to inform ethnographic methods in ways that are particularly useful for researching haptic media practices. This chapter then provides some examples of how the data are always in excess, an overflow resulting from the variability and individuality of participant experience and the difficulty of describing and accounting for personal histories and feelings. Finally, we show how the data can be strategically ‘contained’ in the process of writing for the purposes of publication, thematic framing and knowledge translation.

This chapter reflects on a media studies project exploring Sylvia Plath poetry on Tumblr. The project ultimately resulted in excess digital data, with no conventional publications or research outputs. Now writing 10 years after data collection, I take a storying approach to explore the original research concerns and the research process, thereby locating a reconfigured ‘research event’ that draws together various biographical, social, political and historical factors. I reflect on my evolving understanding of ‘research’, discussing early teaching experiences and postgraduate pathways that partly structured a particular relationship to research. This serves to bridge a discussion about the challenges of the initial process over a decade ago, including the uncomfortable pairing of inexperience among aspiring researchers and institutional pressures to publish. I then discuss the theoretical perspectives that inspire and, in retrospect, offer clarity for the project, given the amount of time passed since data collection and the synergistic relationship between the storying approach, poststructuralist thought and story-focused methodologies. I argue that Tumblr provides unique opportunities for identity negotiation, aesthetic appreciation, data extraction and commodification, which highlights both the creative agency of digital aesthetic curation and self-work, as well as the importance of algorithmic transparency. I also contend that engaging with excess data led to methodologically and theoretically useful insights, challenging assumptions about the temporality of usable data and the ever-changing relationship between art, technology and freedom.

This chapter discusses the challenges and opportunities of integrating big data generated by contemporary museums into data ecology and data fabrics of smart cities. First, it exposes that smart cities could enhance their global reputation, visibility and image by building on closer collaborations with museums. Second, it demonstrates that museums in the 21st century have transformed into hyper-connected cultural hubs, spreading their reach and impact beyond their immediate urban locations. Finally, this chapter discusses creative approaches to data-curation mechanisms that stress the role of museums and cultural heritage sites in supplying data for a more strategic and proactive smart city co-design and management. Specifically, this chapter offers a three-dimensional framework for integrating heritage data in the design of smart city data ecosystems, which includes such components as Data Resources, Data Republics and Data Impacts. Data Resources stresses museum collections’ data and meta-data as a strategic resource to empower creative public data-curation practices to tell meaningful stories about the city and enhance place-making. Data Republics focuses on big data generated by visitors online or on-site as a foundation for evidence-based urban research, design and management, empowering more sustainable, safe and enjoyable tourism. Data Impacts details data-driven methodologies that museums could employ to measure public sentiment and opinion to offer new human-centred indicators to understand the performance of smart cities. This chapter shares a conceptual framework for repurposing museum data within a smart city data ecology to translate the current data excess into data intelligence.

Evaluation and evaluative design aim to assess the impact of programmes, services and interventions. Underpinned by programme logics and theories of change, evaluation aims to assess intervention effectiveness and to determine an intervention’s capacity to produce the intended change and achieve ‘success’. This chapter is focused on evaluative data and the stories that data and its production make (in)visible and the excess data that gets left behind. I document the ways that health interventions use evidence and the shifts in evaluation towards making sense of the complex contexts and systems where interventions are embedded. Taking digital health interventions as an example of a critical contemporary shift in health, I examine the ways digital data is used to offer ‘evidence’ of interventions and how data excess emerges in evaluative research where potentially useful data is not collected or is ignored as seemingly irrelevant. Here, I situate excess in two ways. The first is in relation to the broadening of data that emerges with new digital technologies and what it promises. The second form of excess is data about social life, complexity and practices, which can get left behind when there is a focus on the ‘digital’. I argue that continuing to interrogate the use(s) of digital data is critical for situating health within complex contexts and social practices of everyday life. Excess offers a useful framing to make sense of data and its (non)uses and the implications of such actions in evaluative research.