OII Blog

OII Professor of Society and the Internet Helen Margetts has a piece in the current issue of the ESRC’s ‘Society Now‘ magazine, in a feature that asks ‘Forty years after the first message was sent through the internet we are now connected on a global scale – but do we have a global community?’

I have scraped and pasted Helen’s piece below.

***
‘Social Science of the Net’ by Helen Margetts, Professor of Society and the Internet at the Oxford Internet Institute, University of Oxford [Society Now, Issue 5, Autumn 2009]

The shift of many areas of social and political life onto the internet has major implications for social science research. It provides an opportunity to understand areas of life and society that we have lacked the data and research tools to test before. When have we previously seen the complete transaction history of an organisation – as provided by the freely available edit history of Wikipedia, downloaded in its entirety by a research student at the Oxford Internet Institute?

Daily capture of some of the innumerable online campaigns can provide thousands of ‘joining curves’ of real political mobilisations. Social networking sites, such as Facebook, generate massive amounts of information on social networks. Automatic ‘web-crawlers’ can collect non-obtrusive data on links between and within sites to provide a structural view of any sector – government, commerce and the voluntary sector.

This change is exciting, but it also brings new challenges. At times we have ‘too much’ data. The download of Wikipedia generated a six terabyte database, requiring use of the UK National Grid Service to analyse. Web-crawling provides huge maps of organisational relationships and internal structures, which can be extremely hard to interpret. Thousands of curves showing take-up rates of online petitions or charitable campaigns can be difficult to understand without background knowledge.

Sometimes, we have surprisingly little amounts of data. User data on the internet is the property of the owner of a site, and even if we can obtain it we still know little about where users came from or where they are going. Search engine companies are the custodians of an extraordinary wealth of such data, but they tend not to publish it, let alone share it. There are ethical and legal barriers to data collection. Around 1.5 billion politically-oriented YouTube video clips were viewed during the 2008 US presidential election campaigns, analysis of which would be fascinating for political scientists. But YouTube does not allow the use of automatic crawlers to track download data. Likewise, Facebook operates strict rules which prevent analysis of the thousands of social networks it generates.

Rigorous study of life online is a technologically complex task, continually introducing new research challenges. There are ways around all these challenges though, which mean that studying social and political life is getting more exciting. Social science research increasingly involves borrowing from other disciplines as diverse as physics, computer science and epidemiology, and the development of methodologies such as advanced network analysis, agent-based modelling and experiments. The rich variety of data and insight that these tasks provide is worth the wait and pays back the hope.