For my dissertation, I’ve been working on a way to generate new protest event data using principles from natural language processing and machine learning. In the process, I’ve been assessing other datasets to see how well they have captured protest events.
I’ve mused on before on assessing GDELT (currently under reorganized management) for protest events. One of the steps of doing this has been to compare it to the Dynamics of Collective Action dataset. The Dynamics of Collective Action dataset (here thereafter DoCA) is a remarkable undertaking, supervised by some leading names in social movements (Soule, McCarthy, Olzak, and McAdam), wherein their team handcoded 35 years of the New York Times for protest events. Each event record includes not only when and where the event took place (what GDELT includes), but over 90 other variables, including a qualitative description of the event, claims of the protesters, their target, the form of protest, and the groups initiating it.
Pam Oliver, Chaeyoon Lim, and I compared the two datasets by looking at a simple monthly time series of event counts and also did a qualitative comparison of a specific month.
Michael Corey asked me to post this CfP for a conference “Demography in the Digital Age,” occurring at Facebook the day before ASA (August 15). Note that this is the same day as the ASA Datathon, but if you’re a demographer this looks very cool.
On August 15th 2014, Facebook is sponsoring a conference on data collection in the digital age. Planned for the day before the American Sociological Association meetings in SF, the conference aims to bring together faculty, grad students, and industry professionals to share techniques related to data collection with the advent of social media and increased interconnectivity across the world.
I’m excited to say that Sociological Science, the new general audience open-access sociology journal, has published its first batch of articles. These include a great set of pieces, including one from my collaborator Chaeyoon Lim on network effects and emotional well-being. But the article “The Structure of Online Activism” by Lewis, Gray, and Meierhenrich caught my eye, for obvious reasons.
I’ve got some thoughts on this article, and following the philosophy of Sociological Science of encouraging “ex post corrections/comments over ex ante R&R demands,” here’s my response, which I’m also posting as a formal response on the Sociological Science site.
Brayden King at Northwestern asked me to pass this on.
The Kellogg School of Management at Northwestern University seeks a post-doctoral researcher interested in at least one of the following areas of scholarship: social movements, collective behavior, networks, and organizational theory. We particularly encourage scholars to apply who have advanced quantitative training, programming skills, and familiarity with “big data” methods. The ideal candidate will have a PhD in sociology, communications, political science, or information sciences.
The post-doctoral position will allow the scholar to advance his or her own research agenda while also working on collaborative projects related to social media and activism. The post-doctoral position will be managed by Brayden King and will be affiliated with the Management and Organizations department and NICO (Northwestern Institute on Complex Systems). The term of this position is negotiable.
To apply, please e-mail curriculum vitae along with a brief statement of how your research interests are related to this position to Juliana Steers (firstname.lastname@example.org) with “MORS Post-Doctoral Position” as the subject. Arrange to have two letters of recommendation e-mailed to the same address. Salary and research budget are competitive and includes full medical insurance. Applications are due March 2, 2014.
Northwestern University is an Equal Opportunity, Affirmative Action Employer of all protected classes including veterans and individuals with disabilities.
This is a guest post by Charles Seguin. He is a PhD student in sociology at the University of North Carolina at Chapel Hill.
Sociologists and historians have shown us that national public discourse on lynching underwent a fairly profound transformation during the periods from roughly 1880-1925. My dissertation studies the sources and consequences of this transformation, but in this blog post I’ll just try to sketch some of the contours of this transformation. In my dissertation I use machine learning methods to analyze this discursive transformation, however after reading several hundred lynching articles to train the machine learning algorithms, I think I have a pretty good understanding of key words and phrases that mark the changes in lynching discourse. In this blog post then, I’ll be using basic keyword, bigram (word pair), and trigram searches to illustrate some of the changes in lynching discourse.
This is a guest post by Laura K. Nelson. She is a doctoral candidate in sociology at the University of California, Berkeley. She is interested in applying automated text analysis techniques to understand how cultures and logics unify political and social movements. Her current research, funded in part by the NSF, examines these cultures and logics via the long-term development of women’s movements in the United States. She can be reached at email@example.com.
Computer-assisted, or automated, text analysis is finally making its way into sociology, as evidenced by the new issue of Poetics devoted to one technique, topic modeling (Poetics 41, 2013). While these methods have been widely used and explored in disciplines like computational linguistics, digital humanities, and, importantly, political science, only recently have sociologists paid attention to them. In my short time using automated text analysis methods I have noticed two recurring issues, both which I will address in this post. First, when I’ve presented these methods at conferences, and when I’ve seen others present these methods, the same two questions are inevitably asked and they have indeed come up again in response to this issue (more on this below). If you use these methods, you should have a response. Second, those who are attempting to use these methods often are not aware of the full range of techniques within the automated text analysis umbrella and choose a method based on convenience, not knowledge.
I was pleased to see Fabio Rojas make an open invitation for more female scholars on OrgTheory. Writing for a technically-oriented blog, I’ve been painfully aware of the dearth of female voices expressed here. And as computational social scientists, we should be incredibly wary of the possibility of reproducing many of the same kinds of inequalities that have plagued computer science and tech at-large. We see this when “big data isn’t big enough“, as Jen Schradie has put it, when non-dominant voices are shushed in myriad different ways online, and I fear it when all our current contributors are men. Sociology has gone a long way to open up space for more “scholars at the margins” (a term I’m taking from Eric Grollman and his blog Conditionally Accepted), but there’s still a long way to go.
This is, then, an open invitation for anyone to contribute to Bad Hessian, especially women, people of color, queer people, people with disabilities, working-class or poor people, fat people, immigrants, and single parents. Our doors are always open for guest contributors and new regular contributors. Computational social science ought to be as committed as possible to not only bringing computational methods into the social sciences, but making sure that everyone, especially those at the margins, have a place to speak to and engage with those methods.
2013 was the first full year of Bad Hessian’s existence, so we’re taking stock of what we’ve accomplished in the past year.
We’ve had 37 posts written by the regular crew plus 5 great guest authors.
We’ve had 51,520 unique visits, 39,412 unique visitors, and 70,772 pageviews. Most people are coming from search engines and we’re getting most social media traffic through Twitter.
The five most popular posts of 2013 (written in 2013) were:
- Lipsyncing for your life: a survival analysis of RuPaul’s Drag Race by Alex
- A Final Twitter-based Prediction of RuPaul’s Drag Race Season 5 by Alex
- Cluster Computing for $0.27/hr using Amazon EC2 and IPython Notebook by Randy Zwitch
- RuPaul’s Drag Race Season 5 Finale — Predicting America’s Next Drag Superstar from Twitter by Alex
- Has R-help gotten meaner over time? And what does Mancur Olson have to say about it? by Trey
It was a great year for us. What does 2014 bring? I can think of a few things that’ll probably come up.
- More stats pedagogy
- More IPython
- More social science hackathons and data events
- More discussions of protest event data
- More drag queens (duh)
And I hope more content in general! Is there anything you’d like to see here in 2014? Let us know!
Last month, Mobilization published a special issue on new methods in social movements research, edited by Neal Caren. I was one of the contributors to the issue, submitting a piece borne of my master’s work. The piece is on using supervised machine learning of Facebook messages from Egypt’s April 6th Movement in its formative months of 2008, corroborated by interviews with April 6th activists.
With the emergence of the Arab Spring and the Occupy movements, interest in the study of movements that use the Internet and social networking sites has grown exponentially. However, our inability to easily and cheaply analyze the large amount of content these movements produce limits our study of them. This article attempts to address this methodological lacuna by detailing procedures for collecting data from Facebook and presenting a class of computer-aided content analysis methods. I apply one of these methods in the analysis of mobilization patterns of Egypt’s April 6 youth movement. I corroborate the method with in-depth interviews from movement participants. I conclude by discussing the difficulties and pitfalls of using this type of data in content analysis and in using automated methods for coding textual data in multiple languages.
You can find the PDF here.
The issue is full of a lot of other great stuff, including:
Studying Online Activism: The Effects of Sampling Design on Findings, Jennifer Earl
How Repertoires Evolve: The Diffusion of Suicide Protest in the Twentieth Century, Michael Biggs
Contextualizing Consequences: A Sociolegal Approach to Social Movement Consequences in Professional Fields, Elizabeth Chiarello
A Methodology for Frame Dynamics: Analyzing Keying Battles in Palestinian Nationalism, Hank Johnston and Eitan Y. Alimi
The Radicalization of Contention in Northern Ireland, 1968-1972: A Relational Perspective, Gianluca De Fazio
This is a guest post by Jen Schradie. Jen is a doctoral candidate in the Department of Sociology at the University of California-Berkeley and the Berkeley Center for New Media. She has a master’s degree in sociology from UC Berkeley and a MPA from the Harvard Kennedy School. Using both statistical methods and qualitative fieldwork, her research is at the intersection of social media, social movements and social class. Her broad research agenda is to interrogate digital democracy claims in light of societal and structural differences. Before academia, she directed six documentary films on social movements confronting corporate power. You can find her at www.schradie.com or @schradie on Twitter.
Five years ago, Chris Anderson, editor-in-chief of Wired Magazine, wrote a provocative article entitled, “The End of Theory: The Data Deluge Makes the Scientific Method Obsolete” (2008). He argued that hypothesis testing is no longer necessary with google’s petabytes of data, which provides all of the answers to how society works. Correlation now “supercedes” causation:
This is a world where massive amounts of data and applied mathematics replace every other tool that might be brought to bear. Out with every theory of human behavior, from linguistics to sociology. Forget taxonomy, ontology, and psychology. Who knows why people do what they do? The point is they do it, and we can track and measure it with unprecedented fidelity. With enough data, the numbers speak for themselves.
An easy strawman, Anderson’s piece generated a host of articles in academic journals decrying his claim. The overall consensus, to no surprise, was that the scientific method – i.e. hypothesis testing – is far from over. Most argued as Pigliucci (2009:534) articulated,
But, if we stop looking for models and hypotheses, are we still really doing science? Science, unlike advertising, is not about finding patterns—although that is certainly part of the process—it is about finding explanations for those patterns.
Other analysts focused on the debate around “correlation is not causation.” Some critiqued Anderson in that correlation can lead you in the wrong direction with spurious noise. Others implicitly pointed to what Box (1976) articulated so well pre-Big Data – that science is an iterative process in which correlation is useful in that it can trigger research which uses hypothesis testing.