Is big data a big deal in political science?
By Michael Alvarez
Not a day passes when I don’t see something in the news about big data. Sometimes the stories will be about some interesting new big data application. For example I recently read about the WeatherSignal app that is collecting weather data from smartphones. And of course there has been a lot in the news lately about the big data and privacy, such as news reports that the National Security Agency has been using data to model the social connections of some Americans.
Given all of the discussion in the media about big data, it’s important to ask if the buzz surrounding big data is meaningful for social science, especially in my own discipline of political science. In some ways it seems that big data isn’t fueling a revolution in political science because some political scientists have been using machine learning methods for decades. A quick visit to The Penn State Event Data Project Website shows how Phil Schrodt and his colleagues have long been using machine learning methods to develop and analyze relatively large datasets of political events. University of Georgia’s Keith Poole has been using multidimensional scaling methods for decades to study roll call votes from the US Congress. Both Schrodt and Poole have long been important figures in the development of political methodology, and their work has many of the exact attributes that characterize big data — the use of computationally-intensive techniques to analyze what to social scientists are large and complex datasets.
But while there has been this long tradition in political science for big data-like research, there is no doubt that political scientists encounter larger and larger datasets with increasingly complex structures. Good examples of some of these big data applications appear in the current Political Analysis virtual issue, Big Data in Political Science. In article after article, we see that political scientists are using innovative new big data techniques and methods to collect data and test hypotheses.
For instance, Berinsky, Huber and Lenz show how Amazon.com’s Mechanical Turk can be used for the recruitment of subjects for field experimentation. As field experiments are an important methodology that many social scientists use to test many different behavioral theories, Berinsky, Huber, and Lenz show how large-scale field experiments can be accomplished at low cost. This illustrates how new big data approaches are being applied to develop new means of collecting data for political science research problems.
So is big data a big deal in political science? The answer is neither yes nor no. Political scientists have long conducted research that sure looks like what we now call big data. It is also the case that we are seeing more and more work that uses large datasets and machine learning approaches. My guess is that these trends will continue, since political scientists are seeing interesting new research opportunities with social media data, with large aggregations of field experiment and polling data, and with other large-scale datasets that just a few years ago could not be easily analyzed with available computational resources.
R. Michael Alvarez is a professor of Political Science at Caltech. His research and teaching focuses on elections, voting behavior, and election technologies. He is editor-in-chief of Political Analysis with Jonathan N. Katz. A new virtual issue ‘Big Data in Political Science’ from the journal shows how Big Data tools are being used in cutting-edge political science research, and how political methodologists are contributing to the Big Data revolution.
Political Analysis chronicles the exciting developments in the field of political methodology, with contributions to empirical and methodological scholarship outside the diffuse borders of political science. It is published on behalf of The Society for Political Methodology and the Political Methodology Section of the American Political Science Association. Political Analysis is ranked #5 out of 157 journals in Political Science by 5-year impact factor, according to the 2012 ISI Journal Citation Reports. Like Political Analysis on Facebook and follow @PolAnalysis on Twitter.
Subscribe to the OUPblog via email or RSS.
Subscribe to only politics and political science articles on the OUPblog via email or RSS.
Image credit: Internet Background with Code and Technology World. © kentoh via iStockphoto.