Big Data has revolutionized the way we shop, work and interact. It is also gaining momentum in scholarly research. To show the potential of Big Data analysis in mass communication research, Professor Malcolm Parks from the University of Washington curated a special issue of the Journal of Communication. The goal was to present cutting edge empirical papers and provide a benchmark for research innovation.

Parks argues that we are only at the beginning of substantial Big Data research in mass communication. He also cautions that crunching huge amounts of numbers does not automatically yield interesting results. Big Data cannot and should not do away with thorough research designs and solid hypothesis testing.

Contributions to the special issue span a wide spectrum of research interests even though seven out of eight studies focused on data from Twitter. Eleanor Colleoni and colleagues examined whether Twitter users with similar political attitudes flock together or seek interaction across party lines. Sherry Emery and colleagues looked at reactions of Twitter users to a controversial public health campaign. Fabio Giglietto and Donatella Selva analyzed a complete dataset of tweets spanning over an entire season of TV talk shows to study how audience members interact with television programming.

Three studies focus on agenda setting, two of them with regard to political campaigns. Andreas Jungherr analyzed tweets during the federal election campaign in Germany in 2009 while Chris Vargo and colleagues examined Twitter messages during the 2012 US presidential campaign. W. Russel Neuman and colleagues also studied patterns of issue framing in the social and traditional media in this context (calendar 2012 in the United States) but did not limit their research to the election campaign.

Benjamin Hill and Aaron Shaw take a different approach to Big Data and examine organizational structures of Wikipedia contributors.

Malcolm Parks, who specialises in social networks and interpersonal communications, spoke to the EJO about the challenges of Big Data research. He believes that Big Data leads to “new research questions and new ways of thinking about existing questions” and allows researchers to bring together multiple datasets—datasets of different times, from different places, or gathered at different times. But he warned that it is too early to say whether Big Data will change the way communication is studied.

What is the most important contribution of Big Data research to the field of communication studies?

Malcolm Parks: So far Big Data has yet to make any important substantive contribution to the discipline of communication.  But it is too early in the story to expect substantive contributions.  We are still in a very exploratory stage. Demonstration projects and descriptive analyses dominate the work to date. Little of the current work is likely to have lasting impact, but much of it will contribute to on-going process of conceptual and methodological advance.  As the methods become more available to researchers already engaged with substantively important questions, I am certain that we will see significant contributions involving Big Data.

What are the most challenging limitations for Big Data research?

Parks: In my view the two greatest challenges revolve around validity and access. With regard to validity, there are frequently gaps between the conceptual interpretation of big data indicators and the reality of the indicators themselves.  Access limitations remain significant.  We are often unable to gain access to the most useful data because they are locked behind proprietary or government firewalls.  As a result, we run the risk of making too much of the data to which we do have access.

Where do we go from here? What’s the next frontier for Big Data research?

Parks: We have indeed been in a bit of a frontier mentality with regard to big data.  Sticking with the metaphor, I’d say that it’s time to move big data methods from the frontier to the town where they can be put to work in service of our larger intellectual concerns as a discipline.   Big data does not spell the end of theory or of science, as some have claimed.  Its potential for methodological advance is huge, but only if it helps us address theoretic questions of consequence.  The opportunity continues to grow as these methods diffuse and more of our social activity and media use is captured in data repositories.  But so, too, do the ethical and political challenges of managing access to our increasingly “datafied” selves.

Photo credit: Flikr Infocux