Facebook’s emotional contagion study and the Brave New World of Big Data Research
- Ralph Schroeder
The recent Facebook study demonstrating ‘emotional contagion’ (http://www.pnas.org/content/111/24/8788.full.pdf+html) has generated much debate in the scholarly community and in the media. The study changed the newsfeed of more than 700000 Facebook users, without their knowledge, to see if they would change the kinds of words they posted (positive and negative), and raises the question: how far can researchers go in using social media to understand social behaviour? This question will continue to be discussed, and there will need to be more transparency and stricter guidelines for carrying out this type of research. The ethical and legal aspects of this study are not central to my paper, however. Instead, I ask, why is this type of growth of social scientific understanding of human behaviour worrying at a more fundamental level? The main reason is that big data research promotes the ability to manipulate behaviour (‘manipulate’ in a neutral sense of making us do things). To make this argument requires definitions of ‘big data’, of ‘data’ and of scientific advance. With these definitions in place in the paper, it becomes possible to pinpoint how new data sources from social media are responsible for the rapid take-off in research which can provide more powerful ways to understand peoples’ behaviour – and in this way also to manipulate it (I am not aware of any other paper which has analysed this increased powerfulness and scientificity of big data social media research). Of course, the possibility of such manipulation is quite different in the private sector and in academia. And there may be cases where such manipulation is for the public good. At the same time, this kind of big data research also raises the spectre of the use of scientific knowledge for persuasion that remains hidden and creeps forward, with people unaware of this, or they may embrace being manipulated for their own or for a greater good. In recent years, the world depicted by Orwell in ‘1984’ has been invoked in discussions of the Snowden leaks and other forms of surveillance. I suggest that Huxley’s ‘Brave New World’, in which people readily adapt to a world of more scientifically conditioned behaviour, may be more useful in helping us to think through the implications of big data research using social media. It can be added that the paper is far from being anti- science: As Huxley put it in a BBC broadcast at around the time of ‘Brave New World’, ‘the only cure for science is more science not less’ (quoted in Nicholas Murray’s  biography of Huxley, p.253).
Ralph's full article "Big Data and the brave new world of social media research", published in Big Data & Society earlier this year, is available here
About the author
Economics at Chalmers University in Gothenburg (Sweden). He completed his PhD about Max Weber at the LSE in 1988. His publications include Rethinking Science, Technology and Social Change (Stanford University Press, 2007) and Being There Together: Social Interaction in Virtual Environments (Oxford University Press, 2010). He is also the author of 'An Age of Limits: Social Theory for the 21st Century' (Palgrave Macmillan 2013) and, with Eric T. Meyer, of 'Digital Transformations of Knowledge' (MIT Press 2013). He is currently working on the social implications of e-science and on big data in the social sciences.