The Survey Statistician 22 January 2014 New and Emerging Methods Big Data as a Source of Statistical Information 1 Piet J.H. Daas and Marco J.H. Puts Abstract Big Data is an extremely interesting data source for statistics. Since more and more data is generated in our modern world and is digitally stored, it could certainly be used to replace traditional sources or provide additional information for official statistics. Especially given declines in survey response rates, information gathered from Big Data is an interesting addition. However, extracting statistically-relevant information from Big Data sources is not an easy task. In this paper the current state of the art of research on the use of Big Data for official statistics at Statistics Netherlands is described. The paper is based on the real world experiences of the authors obtained during these studies. The topics discussed are related to Big Data methodology, privacy and security concerns and the skills required for successfully employing Big Data. Introduction Big Data is a term that one hears more and more often at conferences, meetings and seminars. Since its first introduction in 1997, in a conference paper by Cox and Ellsworth (1997), it has really become a hot topic. This is understandable if one realizes that between the introduction of the term Big Data and the present, the world has changed from a ‘data-poor’ environment to a world in which data is abundant (Global Pulse, 2012). This is mainly due to the fact that during this period increasing amounts of data have been generated on the web and by sensors in the ever growing number of electronic devices surrounding us. Because of the ongoing decline in the costs of disk storage this data is no longer thrown away but remains stored. As such, Big Data has the potential to provide information on statistically- relevant populations at high frequency, at a high degree of granularity, and from a wide range of angles, narrowing both time and knowledge gaps. This enables the production of more relevant and timely statistics and can result in proxy indicators that enable richer, deeper insights into human experience than traditional sources of official statistics can (Glasson et al., 2013; Global Pulse, 2012). Anyone who is able to access and analyze Big Data could – potentially – extract meaning from them and gain a competitive edge. This realization has prompted many commercial companies to write white papers and blogs on the huge potential of Big Data. These stories, however, do not always withstand a rigorous scientific analysis and – unfortunately – tend to place the use and potential of Big Data near the edge of the scientific realm. We agree with Glasson et al. (2013), that Big Data has serious potential as it is a very interesting (secondary) data source for official 1 The views expressed in this paper are those of the authors and do not necessarily reflect the policies of Statistics Netherlands.