From Big Data & Society
The ever-expanding usage of social media throughout everyday life offers a critical data resource to social scientists. Though this is increasingly recognised, such data brings with it new methodological challenges in terms of finding ways to analyse what it tells us about social life. Social media provides a form of user-generated data which may be unsolicited and unscripted, and which is often expressed multi-modally (i.e. through combinations of text, hyperlinks, images, videos, music, etc.). Hence, it is important to consider the challenges that this data holds for researchers in terms of rendering them amenable to analysis, and in identifying the sort of research questions that such data might appropriately address. Researchers no longer lack computational tools or theories to help make sense of social media data, yet there remains a paucity of methodologies to make transparent the move from tools to explanations.
This study presents a set of complementary methodologies for undertaking analyses of Twitter data as a socio-technical assemblage, with the emphasis on navigating around and unpicking the factors that construct and constrain the data. Researchers have taken a visual analytic approach wherein visualisations are utilised as tools for forming and pursuing hypotheses rather than results in themselves. The paper demonstrates the value in applying visual analytics to social media research projects by positing four empirical examples as initial steps upon which deeper iterations might be built. The study goes on to demonstrate each of these four approaches ‘in action’, to help clarify how and why they might be used to address various research questions.
In the few years since the advent of ‘Big Data’ research, social media analytics has begun to accumulate studies drawing on social media as a resource and tool for research work. Yet, there has been relatively little attention paid to the development of methodologies for handling this kind of data. The few works that exist in this area often reflect upon the implications of ‘grand’ social science methodological concepts for new social media research (i.e. they focus on general issues such as sampling, data validity, ethics, etc.). By contrast, we advance an abductively oriented methodological suite designed to explore the construction of phenomena played out through social media. To do this, we use a software tool – Chorus – to illustrate a visual analytic approach to data. Informed by visual analytic principles, we posit a two-by-two methodological model of social media analytics, combining two data collection strategies with two analytic modes. We go on to demonstrate each of these four approaches ‘in action’, to help clarify how and why they might be used to address various research questions.
Doing social media analytics
Phillip Brooker, Julie Barnett, Timothy Cribbin
First Published July 8, 2016 Research Article
Big Data & Society