Sloppy Data Floods or Precise Social Science Methodologies? Dilemmas in the Transition to Data-Intensive Research in Sociology and Economics

What are the implications of data-intensive research for social sciences? In an influential editorial, Chris Anderson from the magazine Wired predicted a future where scientific advances would result less from theoretical work, and more from brute force mining on petabytes of data. While evolutionary biology and astronomy are cited as two examples of fields which already entered this new age, it is very uncertain whether the social sciences should feel concerned at all by this prediction. This chapter chooses to examine the situation in two disciplines in particular - sociology and economics - to evaluate to what extent Anderson's claim applies to the social sciences. How is the emergence of new data sources perceived in these fields? Have different types of "big data" had different impacts? (transactional data for sociology, brain imaging data in economics). Is the diagnostic on the desirability of these changes unanimous among sociologists and economists? Comparing the debates and actual paths of research taken in these two cases, we observe that the coming of data intensive research does creates new trajectories renewing existing research practices, but with much more complexity than the sweeping change predicted. This lesson learned for social sciences suggests that a closer look at particular fields from natural sciences could possibly uncover similar, important nuances about the revolution supposedly brought by "big data" science.