Achieving unexpected goals from an ambitious social science project

LongExposurePeople.jpgPart of the magic of bringing people together in multi-team projects is the unexpected things that spark off from their work together — often not at all what was planned or expected, but no less worthwhile for that.

The NeISS project (National eInfrastructure for Social Simulation) brought together creators of social simulation tools with a view to creating an overarching infrastructure with tools and services to look at augmented data, expanding the research possibilities for social scientists across the UK.

Social simulation is a growing field, with applications across the research community from economics to health. Computer models are used to predict how communities might develop and behave, providing useful information to policy makers in both public and private fields. By using a range of sources of information — from census data to the findings of radio surveys — researchers aim to gain new insights into the way people behave and react to events.

In the end, the planned infrastructure turned out to be an aim too far for the NeISS project —coordinating so many disparate teams and developers was impossible within the timeframe and resources of the project. However, some great things did come from the NeISS cooperation and the project’s evaluation and outreach expert Meik Poschen — a research associate at the Manchester eResearch Centre — says that these made the effort worthwhile:

"With all these partners, from such diverse fields, the project was very interesting but quite a challenge to align all the projects and the phases each had reached. But in the end I think it worked. We didn’t create the infrastructure we had hoped for, that was perhaps too ambitious, but all the partners got a lot out of it. Some good thinking was spawned, and a lot of that went on to be reused in other projects."

Neil Chue Hong, director of the Software Sustainability Institute, became involved as technical manager, having worked with many of the project teams through his previous work at OMII-UK. He was pleased with the eventual outputs from NeISS:

"The original goal was to wrap tools from all of the partners and create a single infrastructure allowing social simulation scientists to look at, and model, trends and factors that cause changes in populations. You can use things like Census data, and General Household Survey data, which are very carefully collected, but — in the case of Census data — only gathered every ten years. And then you can augment your simulations with more 'dirty' but up to date data from surveys, direct from the public. It would let researchers pose what if questions to the public and let the public add their own data... Sadly, we never managed to build it! But many parts of the system were integrated, and those secondary outputs have been really well received."

Nine tools have been identified as having come from the NeISS cooperation, along with an increased sense of community and cooperation across the University teams involved. Full information on the tools is available on the NeISS website.

One project that became well known was the Tweet-o-meter, which tracked the volume of tweets coming from different cities around the world. It grew from an initial experiment to visualise the trending Twitter hashtag #uksnow. Tweet-o-meter no longer works, unfortunately, due to changes in the Twitter API, but the team have gone on to develop further tools and projects including one with The Guardian newspaper, analysing Twitter data during the London riots of 2011.

Rob Procter, who recently became Professor at the University of Warwick, working for the Centre for Urban Science and Progress, led the Guardian Twitter work while director of the Manchester eResearch Centre, and continues to work with NeISS project members in his current role. The latest of these, COSMOS (Collaborative Online Social Media Observatory) is building a scalable platform to harvest and analyse social media. Funded by JISC and ESRC, COSMOS brings together partners from Warwick, Cardiff, St Andrews and Edinburgh Universities.

“This work — and the work with The Guardian — wouldn’t have happened without the NeISS project starting it off,” he says.

The Institute’s involvement was essential to the successes that NeISS did see, Poschen says. “Neil [Chue Hong] was really useful for giving broader advice on technologies and holding things together, as he brought a broader view of things to the project. At our bi-monthly meetings Neil was instrumental in getting discussions and proper sessions going, and in developing a diagram of what the whole infrastructure ought to look like, how the partners would connect with one another and so on. Without his input and experience, it would have been much more difficult.”

The NeISS project an was funded by JISC and ran for three years from April 2009 until the end of 2012. It involved teams from the Universities of Leeds, Manchester, Glasgow, Southampton and Stirling, along with University College London.