The final day of IC2S2 kicked off with SAGE Ocean's Katie Metzler introducing the Ethics in Computational Social Science Panel, featuring Dr. Jake Metcalf, Dr. Laura Noren and Dr. Michelle Meyer.
In a year of high-profile media coverage of how our data has been used in unsettling ways, and an increasing body of research suggesting that all this technology might not be making us happier - now is perhaps the ideal time for the computational social science community to come together and discuss these issues. In some ways, 2018 feels like a breakthrough year for computational social science – as David Lazer said following Cambridge Analytica “suddenly everyone knows what computational social science is… but unfortunately they think it broke the world.”
The panel discussed how, though all researchers need to, and do think about ethics, the questions computational social scientists face seem thornier. Most big data used in social research has not been generated by researchers specifically for research purposes and so the usual steps undertaken when collecting data from human participants – such as a formal ethical review, gaining informed consent and de-identification of data may not have taken place, or may be very challenging to do.
In addition to continuing debates about what constitutes private versus public spaces online, questions arise as to the role of informed consent when possible future uses of the data we produce now through our networked digital lives are hard to predict and control. The data we produce are often used in ways that weren’t anticipated when the data were generated or collected and often become valuable only when linked with other data sources, leading to challenges - technical and ethical - around the de-identification and anonymisation of data.
A number of ethical questions related to data access and who holds the data we produce were highlighted. If only companies, and the social scientists working within these companies, have access to really large social and transactional datasets, where does that leave the broader scholarly community? And how do we ensure that the algorithms companies are using to turn our data into profit are fair, accountable and transparent?
How Cascades Grow
The final keynote of the conference was delivered by Lada Adamic, who discussed how different types of networks cascade in different ways.
Her talk featured the evolution and growth of cascades on Facebook, with examples falling into four different categories or protocols: persistent copy, transient copy, nomination and volunteer.
During her talk, Lada highlighted evidence of the behavior of different diffusion protocols - whilst some protocols maximize reach, others maximize engagement, but all reach the same average reproduction number. She also highlighted how the social cost and individual effort and tie strength also played a role in differing diffusion patterns.
It was a fantastic few days, bringing together a diverse range of people from across academia, industry and government. We heard and learnt from some of the world's leading Computational Social Scientists and gained insight into a range of different inter-disciplinary research projects. You can check out some of the highlights from Twitter here.Huge thanks to the organizers for putting on such a great event.
It has been announced that IC2S2 2019 will be held in Amsterdam, NL - we hope to see you there next year!