Guest Editorial: What’s so big about Big Data? Finding the spaces and perils of Big Data

Abstract

For scholars, entrepreneurs, and technology writers, 2012 was a watershed year for ‘‘Big Data’’—here referring to massive datasets produced through the aggregation of crowdsourced, social, and other digitally available data (Mayer-Schonberger and Cukier 2013; Rasmus 2012; Nicole 2012). The ‘‘year of Big Data’’ (Rasmus 2012) saw a drastic rise in the use of Big Data and its related methods and technologies in fields as diverse as marketing (see: Baker 2013), healthcare (see: Cerrato 2012), international development (see: Letouze 2012), humanitarianism (see: UN OCHA 2013), and national funding agencies (see: National Science Foundation 2012) (Thatcher 2014). More recently, geographers studying technology have also turned their gaze to ‘Big Data’ and its variegated spatialities (Barnes and Wilson 2014; Batty 2013; Goodchild 2013; Kitchin 2013, 2014; Crampton et al. 2013). While much data may always have been big (Dalton and Thatcher 2014b; Kitchin 2013, this issue), and the manipulation and analysis of spatial data may have a long and complex history within geography (Goodchild 2006), the ability to rapidly aggregate and analyze previously unheard of combinations of data has led to an increased focus on the relations between data and knowledge production (Boellstorff 2013; Thatcher 2014). For many, the panacea to diverse social ills has become larger data sets and quantification, in what could be construed as ‘‘naive empiricism’’ (Taylor 1990, 212). At worst, this view can devolve—as it has in at least one well-known case— into the ‘‘end of theory’’, wherein numbers have come to ‘‘speak for themselves’’ (Anderson 2008). As government, private industries, and academic researchers all rush to embrace Big Data, some scholars have pushed back, calling into question the ‘‘purely data-driven approach’’ (Kling and Pozdnoukhov 2012, 483) and its epistemological, economic, and political commitments (Batty 2012; Boyd and Crawford 2012; Burgess and Bruns 2012; Richards and King 2013; Dalton and Thatcher 2014b). This special issue responds and contributes to those debates by calling attention to and exploring the continued importance of ‘‘small data,’’ within the context of Big Data’s continual rise to prominence. Here we conceptualize ‘‘small data’’ as datasets and data production and analysis methodologies that are limited in size and scope relative to Big Data, but may contain rich, contextualized data that has been produced with a particular purpose in mind. Growing out of a session at the 2012 AAG entitled ‘‘Whither Small Data? R. Burns Department of Geography, University of Washington, Smith Hall 408, Box 353550, Seattle, WA 98195, USA e-mail: burnsr77@gmail.com