Data journeys in the sciences

Abstract

What is the point of data in research? Philosophers and methodologists have long discussed the use of data as empirical fodder for knowledge claims, highlighting, for instance, the role of inductive reasoning in uncovering what data reveal about the world and the different ways in which data can be modelled and interpreted through statistical tools. This view of data as a fixed, context-independent body of evidence, ready to be deployed within models and explanations, also accompanies contemporary discourse on Big Data – and particularly the expectation that the dramatic increase in the volume of available data brings about the opportunity to develop more and better knowledge. When taking data as ready-made sources of evidence, however, what constitutes data in the first place is not questioned, nor is the capacity of data to generate insight. The spotlight falls on the sophisticated algorithms and machine learning tools used to interpret a given dataset, not on the efforts and complex conditions necessary to make the data amenable to such treatment. This becomes problematic particularly in situations of controversy and disagreement over supposedly “undisputable” data, as in the case of current debates over the significance of climate change, the usefulness of vaccines and the safety of genetic engineering. Without a critical framework to understand how data come to serve as evidence and the conditions under which this does or does not work, it is hard to confront the challenges posed by disputes over the reliability, relevance and validity of data as empirical grounds for knowledge claims.