Digital data is not static. Rather, digital data is continuously on the move – passing rapidly through various sites of data practice where different people and technologies are involved in its (re)production, (re)processing and (re)use. This is why social scientists talk about data ‘circulation’, the ‘social life’ of data, and ‘data journeys’. As such, it is important to consider the ways in which digital data is made to move from one place to another.As Jo Bates reminds us, “clearly data do not move of their own accord” (2018, p.413). So how, where, when and why does a piece of digital data go once it has been generated?
A key part of our investigations into how schools ‘do’ data will be describing data journeys. The notion of ‘journey’ draws attention to the ways in which data travels, as well as the ways that it might stop temporarily in a place, be blocked or simply slow down. In this sense, it is important to remember that data does not ‘flow’ fluidly and effortlessly. Indeed, describing a ‘journey’ is an explicit acknowledgement of the sites that data ‘stop off’ at, and the actors that is meets along the way (Dalton and Thatcher 2014). In addition, is the need to consider the range of ‘data frictions’ that might be encountered – impediments associated with infrastructural inefficiencies, regulatory restrictions, and the various data cultures that shape the sites where the data is processed and used. Taking account of all these aspects of data journeys therefore allows us to explore the “contingent and contested social practices” (Bates et al. 2016) that shape the production and interpretation of all digital data.
Critical data studies is not the first area of scholarship to conceptualise data in this manner. Mapping the movement of data is a long-standing interest of the information sciences. Similarly, scholars working in infrastructure studies and documentary studies have honed the art of ‘inverting’ knowledge infrastructures (Bowker 1994) – i.e. looking beneath the surface and exploring the social and relational nature of how data and information is circulated and conveyed. Latterly, the STS-inflected trend for ‘trace ethnography’ has also renewed interest in empirically following digital data where-ever it travels.
Our interest in ‘data journeys’ therefore continues these traditions – as Bates et al (2016) put it, seeking “to better situate data across interconnected sites of practice distributed through time and space, drawing attention to the movement of data between these sites” (p.2). In this sense, the notion of ‘data journey’ also prompts us to consider the places where data could go (but does not), as well as the places where it cannot go. Accounts of any ‘data journey’ will therefore detail the slow-downs, breakdowns, and end-points of these movements of data, as well as the varying speeds and timings of these movements. Such descriptions also shed valuable light on the ‘mutability’ of digital data – i.e. occasions when digital data is appropriated, manipulated and adapted for different purposes by different actors.
So, how might we practically research the journey of digital data? Bates and colleagues (2016) offer a useful set of methodological proposals for “illuminating the socio-material life of data as they travel between and through different sites of data practice. Their approach is based on following pieces of data through multiple interconnected organisations and projects within and across knowledge infrastructures” (p.3). In brief, this involves in-depth ‘desk research’ to produce rough mappings of the main sites of data practice that a piece of digital data might journey through. These sites are then visited by researchers (either in-person or online) as locations for interviews, field observations and documentary analysis of relevant policies and protocols.
When replicated across successive sites of data practice, these sustained qualitative inquiries can quickly amass detailed accounts of the production of a data entity. This will encompass the cultural practices and conditions behind the initial production of the data entity (e.g. the reasoning underlying its timing, units of measurement and so on), as well as the specific material forms in which the data is subsequently represented (e.g. types of file format, platforms that it is passed through, databases that it is stored in). Indeed, some of the most interesting techniques to utilize here are software-based and code-driven inquiries of data-logs, meta-data, user interfaces and data system architectures.
That said, throughout these investigations, a stalwart method continues to be oral history interviewing with key participants. As Bates et al. (2016, p.5) reason:
“[using] oral history interviewing in our conversations with participants, we [a]re able to draw upon their memories of the development of the infrastructure in order to construct an evolutionary and dynamic picture of the life of data which emphasises key moments in the development of data journeys and practices” (p.5)
All told, all these in-depth qualitative approaches can be considered produce rich and ‘thick’ descriptions of a data journey in the best traditions of ethnographic research. We hope that such inquiries will provide valuable insights in the broader nature of datafication and its ramifications in the context of a school and/or school system.
[notes on Bates, J., Lin, Y. & Goodale, P. (2016). Data journeys: Capturing the socio-material constitution of data objects and flows. Big Data & Society, 3(2), 2053951716654502]