Building an apparatus: Reflective and diffractive readings of trace data

Publication Type:





When people interact via information systems, data are captured by the systems about the interaction as a side effect. These data, referred to as trace data, are increasingly available and interesting for research. In a sense, these systems form a kind of research apparatus, and like all advances in instrumentation, open new areas of study with great potential for discovery.

At first glance, such “big data” would seem to be most suitable for a quantitative and positivist research approach. However, we argue that considering such systems from a socio-material perspective offers insight into the challenges in analyzing trace data and suggests a rethinking of both quantitative and qualitative approaches to trace data. Building on Haraway and Barad’s distinction between reflective and diffractive methodologies, we discuss and illustrate the strengths and weaknesses of these two approaches. A reflective methodology considers trace data as reflecting pre-given objects, people and practices. A diffractive methodology traces the ripples emerging from interference between trace data, information systems, users, designers and researchers.

Drawing on longitudinal study of citizen science practices that uses trace data, the paper illustrates the consequences of each methodology. We show that a diffractive methodology allows researchers to account for not only the socio-material dynamics of digital trace data but also the temporal dimension of online practices, directing the researcher’s attention to how the apparatus configures and reconfigures not only trace data, but also the information system, users, system designers, and researchers.