By Fidelia Dake, faculty member at the Regional Institute for Population Studies at the University of Ghana
Coming from a social science background, I have had very limited exposure to data science. I was therefore excited to learn about the emerging field of computational social science and the Summer Institute in Computational Social Science (SICSS) presented the right opportunity. I applied to the 2019 SICSS and I was accepted for the Cape Town partner site. I went in not knowing what to expect but by the end of the first day I knew the experience at the two-week Summer Institute was going to be truly worthwhile.
The institute combined lectures (delivered by researchers from different disciplines), practical sessions and group work on different research projects. The lectures covered topics such as big data, ethics of sourcing and using big data, and different methods of analysis in data science.
A number of things stuck with me throughout the Summer Institute. The organizer of the Cape Town partner site, Dr. Visseho Adjiwanou continually challenged us to think about new ways in which conventional social science research can be conducted leveraging on the opportunities that big data in the digital age present. He also pointed out the possibility of using these kinds of data to solve real problems facing the African continent. Professors Matt Salganik and Chris Bail also emphasized that while big data present opportunities, it also presents challenges, especially with ensuring privacy and confidentiality. They further emphasized that the highest ethical standards need to be applied in computational social science research.
I left Cape Town with three key lessons from participating in SICSS:
Firstly, SICSS taught me to think differently about what data are and the different types and forms in which data for social science research can be obtained. For example, data from social media such as Twitter, Facebook and Instagram among others can be independent and complementary data sources for social science research. Additionally, the increasing use of mobile phones and greater access to internet in Africa presents opportunities for new ways of collecting data for research. A typical example is the study conducted by Blumenstock et al. (2015) in Rwanda where mobile phone metadata was used in predicting poverty.
Secondly, I learnt about different analysis techniques including web scrapping, topic modelling and machine learning among others. I got a better understanding of the different applications of machine learning including classification and prediction from working with other participants on our group project that focused on the “Evaluation of machine learning methods for predicting the risk of child mortality in South Africa”.
Thirdly, the application of different methods of computational social science to the varied fields of research that were presented by guest speakers got me thinking about new ways in which I can conduct my own research. Meghan Bruwer’s presentation on working with big data in the field of transport engineering for example aligns with my research on physical activity environments.
All in all, SICSS was a really great experience for me. The mix of early career social scientists and data scientists provided a great learning opportunity for all participants. Working together on group projects with other participants fostered the exchange of knowledge and ideas and building networks and collaborations that will last long after the two weeks we spent together. I am looking forward to sharing my SICSS experience with other researchers. I encourage other early career researchers to consider learning about computational social science. Learning materials and other resources are freely available online.
Fidelia Dake is a faculty member at the Regional Institute for Population Studies at the University of Ghana. She holds a Doctor of Philosophy and a Master of Philosophy in Population Studies. She also holds a Master of Science in Global Ageing and Policy and a Bachelor of Science in Nutrition and Food Science. Her research focuses broadly on health demography, public health, and international health and development. Her interests include nutrition and physical activity, obesity and non-communicable diseases, socio-environmental determinants of health, urban health, health statistics and health-financing.