Boston APSA 2018
Research Cafe for APSA 2018: Computational Analysis of Text and Big Data: Advances in Comparative Politics and Beyond. Comparative Politics Section.
Recent advances in the computational analysis of text, together with the exponential rise of online, including social media data repositories, have opened up exciting possibilities for the rapid growth in data available to political scientists to answer old and new questions of interest to the discipline. In this round-table, participants will outline some of the ways in which they have managed to bring these new tools and possibilities to their own research, and will seek cross-pollination between the techniques they study, and the fields they represent. Human rights, protest, propaganda, elections, political violence, deliberations are among the areas that are being transformed by the creative appropriation of techniques developed by computer scientists and digital humanities scholars to retrieve information from large bodies of online text. We will especially aim to open up dialogues with scholars relying on more more traditional data gathering methods. We will ask how the new approaches fit into current empirical approaches in different fields in political science, and wether innovative solutions can be deployed in tandem with traditional data gathering techniques to further research by expanding the quality and quantity of data available to scholars.
Nikolay Marinov, University of Mannheim
Marinov can speak about a number of applications of text analysis to issues in comparative politics and international relations. One is a an original dataset of economic sanctions, and discussion of economic sanctions, generated by applying machine-learning methods to U.S. Congressional documents and Presidential statements. A related one is using a variety of techniques, including named-entity recognition, to understand when American policy-makers discuss other countries’ elections, leaders, human rights and commitments to international treaties. A third application concerns how to link up all this information to information available on the web on countries’ elections, including information on who ran and on how competitive the election was. The result is a new body of data, uniquely suited to answer questions of interest to political scientists of all stripes, including questions about American foreign policy, sources of influence on other countries’ human rights, and on the electoral outcomes we observe around the world.
Anita Gohdes, University of Zurich
Gohdes can speak about using supervised machine-learning for text-classification for different types of data projects. An advantage of using supervised methods is that researchers can establish a clear codebook that is driven by theoretical concepts decided on a priori. For example, supervised ML can be used on qualitative accounts of individual instances of human rights violations to establish more fine grained measures of violence in contentious environments. Gohdes used supervised ML to classify 60 thousand individual records of fatalities in the Syrian conflict to establish whether individuals were killed in a targeted or indiscriminate way. In a different project, she and co-authors used a small hand-labelled training set of social media posts to classify all Twitter and Facebook posts shared by world leaders. While supervised methods have a lot of advantages, their performance is dependent on a number of important factors that will be subject of discussion in the research cafe.
Rochelle Terman, Stanford University
Rochelle can discuss her experiences applying computational tools and techniques to issues of culture, norms, and identity. These topics areas have historically relied on qualitative and/or critical methods, but recent advances in computational methods have provided new opportunities for engagement. Rochelle will discuss her usage of text-as-data methods, webscraping, and other techniques to examine American media coverage of women's rights and gender norms around the world. She will also discuss her experience as a data science instructor to students across the social sciences and humanities, who apply these techniques to a range of substantive topics using a variety of empirical and epistemological approaches.
Walter Mebane, University of Michigan
Mebane can speak about using Twitter to extract observations of election incidents by individuals across large elections. Automated machine classification methods in an active learning framework have so far been used in the 2016 election in the United States (including primaries, caucuses and the general election) to classify Tweets for relevance and by type of election incident. Even though humans use both text and images to decide how to label Tweets, the machine classifiers currently use text only. Mebane will discuss ongoing work to build neural networks that use both text and images. The project also uses a database of Tweet and user information to support analyzing the data. For example, the user database is useful for filtering out both bots and users identified as bad actors created by Russia, as well as for developing attributes of individual users and of networks of users. For the general election we develop from 16.5 million raw Tweets hundreds of thousands of incident observations that occur at varying rates in different states, that vary over time and by type and that depend on state election and demographic conditions.
Pamela Ban, Harvard
Pamela will discuss how she uses text-as-data methods on congressional text sources to shed new light on theories of congressional politics and organization. Much of the existing empirical work on Congress revolves around using roll call voting data or Congressional Record speech data, which largely limits empirical analyses to the floor-voting stage. Pamela will discuss how she uses new text datasets of committee speeches and committee reports to open up the black box of the congressional committee stage. In particular, using these text sources, she constructs measures of disagreement during the committee stage and investigates how this disagreement affects committee decisions and subsequent floor voting. She explains how incentives present in a strong committee system can lead legislators to deviate in their voting and contribute to bipartisanship. More broadly, she will discuss how using text-as-data can help us understand deliberation processes in Congress.
I will be a discussant for the following proposed panel.
APSA panel: Contentious politics, regime change, and democratization
1. Diversity of Resistance Networks and Post-Campaign Democratization
Jessica Maves Braithwaite (Arizona, non-presenting coauthor), Charles Butcher (NTNU, non-presenting coauthor), Jonathan Pinckney (NTNU, presenting author)
A growing body of work examines how dynamics of violent and nonviolent resistance campaigns influence democratization processes, with nonviolent campaigns commonly thought to be more successful at promoting post-campaign democracy. However, we are often forced to make assumptions about these resistance campaigns’ composition, in particular that nonviolent movements are more inclusive and diverse than their violent counterparts. However, there is considerable variation in terms of broad societal support for both violent and nonviolent campaigns. We expect that campaigns featuring greater levels of diversity in the types of social organizations participating in resistance will be more likely to see the evolution of inclusive, multiparty post-conflict political institutions – regardless of the dissent strategies employed during the campaign itself. We reexamine the prospects for democratic advancement following anti-government campaigns by using novel data on social organizations that mobilize in support of rebel groups and mass nonviolent movements in post-Cold War African states.
2. Not a one way street: Coup d'etats, civil society mobilization, and regime change
Marianne Dahl (Peace Research Institute Oslo)
Kristian Skrede Gleditsch (University of Essex, Peace Research Institute Oslo)
Coups and civil military relations are an important source of political change, but its impact on transitions to democracy has generated much debate. Whereas some see coups in autocracies with intra-elite divisions and prospects for military defection as forces that can promote transitions to democracy, others emphasize how coups remain more likely to usher in new autocratic regimes, where new elites often consolidate power through repression without meaningful political reform. We distance ourselves from optimists who argue that coups in general (both failed and successful) tend to generate incentives for leaders to democratize. We argue that the expected effects of coups is not inherent to the attempts per se, but must be considered relative to other forces that help support democratization, such as popular protest from below. Coup attempts are more likely to spur democratization when they occur in the presence of significant popular dissent. In the absence of dissent, new autocratic rule is more likely. Our analysis of the impact of coups and dissent on subsequent political provides strong evidence consistent with our argument.
3. Social actors and the dynamic of mass resistance campaigns
Sirianne Dahlum (Varieties of Democracy Institute, Gothenburg University)
This study investigates whether the social composition of mass resistance campaigns conditions their choice of tactics and regime-responses. A growing literature investigates the determinants of campaign dynamics, pointing to factors such as the military’s behavior and the structure and size of the campaign. This paper argues that the strategic behavior of campaigns and regimes is also influenced by the social class profile of protest campaign participants, as social background shapes the resources and interests of protesters. I hypothesize that particular urban-based groups – such as urban middle class and public employees – have the resources and interests that facilitate nonviolent strategies. Moreover, social groups with large international networks – such as industrial workers and urban middle class – should be more likely to elicit foreign support. To assess this, I utilize novel data on the social composition of resistance campaigns, recording the involvement of groups such as the urban middle class, industrial workers and public employees. The paper also investigates whether the class composition of opposition movements affect the regime's decision to use violent repression
4. Luke Abbs. United we Stand, Divided we Fall: Cleavage Structures and the Onset of Nonviolent Resistance
Successful nonviolent resistance often hinges upon the ability of movements to mobilise large and diverse numbers of people against the government. Case literature suggests that social divisions are therefore a key challenge for nonviolent movements as divisions increase participation barriers and undermine intergroup coalitions. Yet how different types of ethnic cleavages impact the emergence of nonviolent action remains poorly understood. Moving beyond assumptions that ethnic groups are monolithic, I argue that cleavages within and between ethnic groups generate divisions that constrain nonviolent mobilisation, while cross-cutting religious cleavages linked to pre-existing religious networks facilitate nonviolent action. I examine this relationship using new data on religious cleavages (EPR Ethnic Dimensions Dataset) that cut both within and across ethnic groups. These cleavage structures are explored across a global analysis of nonviolent campaigns onsets from 1946 to 2013 (Major Episodes of Contention), and the onset of mass nonviolent events in Africa from 1990 to 2008 (SCAD). From these analyses, I find evidence that intra and interethnic cleavages undermine the ability of movements to mobilise, while cross-cutting religious cleavages facilitate nonviolent resistance.