Social Media for Social Science Research (Doctoral Research Methods Workshop Series, Part 2)

About

In the space of a few years social media platforms such as Twitter and Facebook have come to occupy an important place in social science research. Their astonishing uptake around the world, and the relatively open, structured nature of the data they produce, offer the tantalizing possibility of studying a wide variety of different social research questions in a way that previously would have been impossible. Already a number of interesting studies have used data from such platforms to shed new light on diverse concepts such as information diffusion, political behaviour and mobilization in social movements.

Despite this promise, the techniques by which such data can be accessed and used remain outside the skill set of many social researchers, whilst the limitations and difficulties of using such data are poorly understood. A variety of new theoretical, methodological, ethical, legal and practical questions face researchers who want to use social media data, for which existing undergraduate and master’s programs are unlikely to prepare them.

This day long workshop proposes to address some of these questions. It is designed specifically for PhD researchers who would like to use social media data in their research but aren’t sure where to start; though researchers already using such data are also welcome to attend.

Topics to be Covered

There are two main outcomes from the workshop, one theoretical, and one practical. In particular, those taking it should leave with:

• An understanding of the theoretical, methodological, ethical and legal challenges involved in using social media data in social science research. We will, for example, tackle questions such as biases in user base, sample sizes, notions of informed consent and the legal intricacies of storing personal data created by private companies.
• An idea of how to start collecting social media data using the Python programming language. This is a short session and complete beginners should not expect to become experts in computer programming straight away. Nevertheless researchers should leave with a basic programming environment installed on their laptop and an idea of how to develop these skills further if they want to.

Who Should Take This Course?

This workshop is designed specifically for “beginners” in the field of social media data: people who perhaps have an idea for how they might want to use such data in their research, but aren’t sure of how to go about collecting it. People already collecting such data may be interested in the parts of the workshop discussing theoretical, methodological, legal and ethical challenges involved in its use, and are welcome to only attend that part of the session if they wish (there is no upper limit to how many people can attend the first part of the workshop, but please get in touch with the workshop organiser if you would like to only attend this first part).

Researchers are required to bring a laptop computer to the session so they can try out the practical programming examples in class.

Format

The workshop is divided into three parts:

1. A lecture and discussion giving an overview of social media data in the social sciences, highlighting some of the possibilities, and discussing some of the theoretical, methodological, ethical and legal challenges
2. A practical workshop session where researchers will be taught how to start to collect and store social media data through an API in the Python programming language
3. An optional afternoon lab session where we can continue the work begun in session 2, try out some extra exercises, or discuss specific issues researchers may have.

Optional readings

These readings may provide some food for thought on the subject but are not a requirement for the workshop:

• Savage, M., and Burrows, R., (2007), The Coming Crisis of Empirical Sociology. Sociology, 41(5): 885.
• Lazer, D., Pentland, A., Adamic, L. et al. (2009), Life in the network: the coming age of computational social science. Science, 323(5915), 721–723.
• Bruns, Axel and Stieglitz, Stefan, Towards more systematic Twitter Analysis: metrics for tweeting activities, International Journal of Social Research Methodology. 16(2):91-108.
• Morstatter, F. J., et al., (2013) Is the Sample Good Enough? Comparing Data from Twitter’s Streaming API with Twitter’s Firehose. ICWSM ‘13. AAAI Press.

This is the best practical text book in the area at the moment. Again, not a requirement, but may be useful for people who wish to keep learning after the course: Matthew Russell, Mining the Social Web, Second Edition. 2013. Pasadena, CA: O’Reilly Media.