Big Data: Methods for Collection and Analysis

Today there is a dual revolution going on in the social sciences. First, more and more daily life leaves behind digital traces which are archived in various databases creating new data resources for social scientists to study social and political life. Alphabet’s Eric Schmidt once claimed that we produce as much information in two days as we had in all of human history up through 2003 (Siegler, 2010). Whatever the veracity of the claim, the accelerating rate of information production and its utilisation in all aspects of political and social life is undeniable (Crozier, 2010, 2012; Ekbia et al., 2015). Second, innovations in computational tools are increasingly making these data accessible to academic researchers through the creation of various libraries and packages which extend popular data analysis platforms, such as NVivo, and programming languages, such as Python and R, to facilitate the collection and analysis of these data. ‘Big data’ is a popularised term that has come to refer to the collection and analysis of digital traces. Similar terms to characterise these methods include ‘computational social science’ (Bankes et al., 2002; Conte et al., 2012; Alvarez, 2016), ‘data science’ (Loukides, 2011; Baesens, 2014; Grus, 2015) and ‘digital methods’ (Rogers, 2009). These methods are being used by political scientists (King et al., 2013; Barberá, 2015; Freelon et al., 2015; Jensen, 2016; Jungherr, 2015), sociologists (Ackland, 2013), communication researchers (Lewis, 2015; González-Bailón and Wang, 2016) and literary scholars (Ramsay, 2011; Jockers, 2013).