Xennials are the demographic cohorts born in the late 1970s to early 1980s that are described as having had an analogue childhood and a digital adulthood. In this book, I want to use this neologism as a metaphor to describe the intellectual biography of social research scholars who were academically ābornā before the data deluge (Anderson 2008), when social science was still analogue, and then adapted to the digital ārevolutionā in research methods.
As many other Xennials, I started my academic education when Apple entered the consumer portable market and completed it just before big data emerged as a buzzword. As such, I take pride in having been able to undertake social research before and after internet-generated data became a regular appearance in the research design of social research projects and PhD dissertations.
By using the term Xennials, I do not mean literally that the experience of moving between analogue and digital social research should be seen a prerogative of the micro-generation born exactly between 1977 and 1983. There are illustrious examples of more senior scholars who also personify research agendas that are grounded in long-standing sociological interests and have subsequently turned to consider the adoption of computational methods in addition to established social research methods.1
There have been many ways to explore the tension between analogue and digital methods (see Beaulieu 2016 for a review) but none of them explore this in a processual way. The purpose of this book is to explore the tension between analogue and digital as part of an evolving research programme and to explore the sequencing of methods within it. The book also responds to a growing demand to place digital research more organically in the context of existing ways of doing social science (Savage 2015: 297). Quoting C. Wright Millsāan author highly regarded by digital research scholars (see Edwards et al. 2013; Savage 2015)āwith this book I also want to call for sociologists to address connections between personal troubles with digital data analysis and more public issues regarding the role of social science in the digital age (Mills 1959).
Most of the authors who have recently written about digital social research fail indeed to rememberāor intentionally omit from their accountsāwhat they were doing in their research before the digital as if it was an incoherence.2 This intellectual oblivion contributes to construe the digital as the bearer of epochal change in social research, an epistemological posture that in itself contributes to make the very notion of digital research unsettling for most sociology scholars, historically not prone to a rapidly moving research front (Collins 1994). As we shall see, social data science should be better placed within the genealogy of the long-term disciplinary relations between phenomenological sociology, expertise in computer science associated to digitalisation and the narrative positivism (Abbott 1992) linked with the use of statistics in social research.
The Disunity of Social Data Science
Technical boosterism is not the only reason for an uptake of digital data analysis in social research that remains on a smaller scale than some early visions initially anticipated (Halfpenny and Procter 2015: 5; Lazer and Radford 2017: 20). The other main reason is the continued ambition in social data science scholarship to speak to multiple audiences at once as in an attempt to define specific parameters around the (inter-)discipline. This is apparent either in monographies and edited collections aimed to address all issues and topics that could be incorporated under a sociology of digital technology (Lupton 2014; Orton-Johnson and Prior 2013) or in works explicitly written for both the social science and the data science audiences (Salganik 2018; Veltri 2019). Some work of the former category is theoretically sound. But it remains conceptual. The conclusions are convincing but rarely put into practice. On the other hand, there are practical guides, where authors provide a wide variety of example-driven accounts that target the skeptics and the enthusiasts, the social scientists as well as the data scientists.
My argument is that both approaches reproduce a rather benign view of a unified social data science. Paradigms are so many and their combination so unique across big data sciences (Bartlett et al. 2018) as well as within social data science that any attempt to construct the (inter-)discipline as a straightforward project is bound to remain conceptual and solidify disciplinary borderlands.
This volume follows up recent conceptual scholarship in digital sociology (for an excellent example see Marres 2017) with a more āconfidentialā, first-person narrative about research methods in this particular form of contemporary interdisciplinarity (Sandvig and Hargittai 2015: 2). It also goes to suggest that a degree of digressions and transgressions (Rheinberger 2011) under the umbrella of social data science is a productive feature.
The book provides a first-hand account from the perspective of an ethnomethodology-trained qualitative social scientist that over time got to approach social problems by using alternative methods also including digital methods. As such, the book is written to intervene in internal discussions within the social sciences about possibilities for collaboration with data science. It is addressed specifically to scholars who, like the author, routinely engage in critical sociological analysis of the digital workplace and find it easier to treat the digital as an object of study. It describes how it happened that the transformation of the workplace taking place in the 10-year arc of a career spent doing fieldwork in the IT sector led to progressively question existing methodological presuppositions in social research and āembraceā new forms of data and methods.
If the goal is to act upon a situation in social data science that Bartlett et al. (2018) describe as ā98% computer scientists and 2% sociologistsā, the editorial politic of producing introductory books on data analytics for social scientists might not be the only possible solution. Another, arguably more promising approach is to consider the theoretical, epistemological and ontological sensibilities that might be involved in a commitment to digital data analysis. Even more important for an audience of empirical social researchers influenced by symbolic interactionism and ethnomethodology is to provide a narrative that is concrete and accountable. This work is a succinct effort in this direction. It targets a specific audience, however large. It does so from a partial, but hopefully well-situated perspective: the first-person perspective of how a particular research agenda has been conducted in the field.
A Processual Perspective on Social Data Science
Differently from most of the rapidly burgeoning literature on digital social research, the bulk of the book is about how a social scientist can gradually get to appreciate digital research methods. The most part of what you will read describes the progression of issues, methods, questions and approaches through which the always coming crisis of empirical sociology (Gouldner 1970; Savage and Burrows 2007) can be approached without necessarily jumping straight onto the digital bandwagon. This aligns with the idea that the focus on social science ideas in the context of the changing character of social phenomena should be as much on what endures as it is on what changes (Housley and Smith 2017).
Continuing w...
