Recent scholarship in international relations, (international) political economy, and comparative politics has witnessed a surge of mapping and studying (legal) documents of international and domestic institutions. New methods have simplified and accelerated the process of data collection. The sheer number of large N / comparative data sets is vastly increasing. Text analysis may complement and corroborate these various mapping projects. Although scholars have long recognized the potential of analyzing large bodies of texts and sometimes even spoken words for the study of international and domestic institutions, many have shied away from such analyses due to the enormous costs associated with them. Automated, computer-assisted text analysis promises to remedy these costs for both the collection and analysis of large N textual data.

The workshop consisted of two parts. In the first part, the participants discussed current scholarship and projects that make use of textual data in the study of international institutions. In the second part, we turned to recent methodological achievements and developments in text analysis (‘text as data’). Participants discussed available methods and tools, and whether and how these can be used to analyze large N data sets in the study of international institutions.

One set of papers explored the usability of text analyis for study of Preferential Trade Agreements (PTAs) and Bilateral Investment Treaties (BITs). KFG Senior Scholar Soo Yeon Kim (National University of Singapore) analyzed text similarity. Andreas Dür and Lisa Lechner (both Universität Salzburg) compared the manual and computational estimates of latent traits in PTAs, and Wolfgang Alschner and Dmitriy Skougarevskiy (Graduate Institute of International and Development Studies) presented a paper that situated the newly signed TPP in the broader investment treaty universe. Finally, Tomer Broude, Yoram Z. Haftel (both Hebrew University of Jerusalem), and Alexander Thompson (Ohio State University) compared the regulatory provision in BITs based on text analysis.

In the second set of presentations, the workshop participants presented papers that dealt with different types of international organizations. Dan Honig (Johns Hopkins University) and Alexander Kentikelenis (University of Oxford & Harvard University) presented their work on the influence of the Executive Board of the International Monetary Fund (IMF), while Tobias Hofmann (University of Utah) outlined a project on informal precedence in the World Trade Organization (WTO) and the potential applicability of automatized text analysis. Anja Jetschke and Patrick Theiner (both Universität Göttingen) presented their work on diffusion processes between regional organizations. KFG Postdoc Fellow Mor Mitrani analyzed and compared globalization discourses of member states in several international organization.

In the final section, Iñaki Sagarzazu (University of Glasgow) and Heike Klüver (Universität Hamburg) presented papers that provided more insights into potential methods and tools for studying international institutions. Iñaki Sagarzazu compared the issue selection in parliamentary debates in Spain, and Heike Klüver’s talk revolved around various ways of measuring interest group influence in the European Union.

The workshop concluded with a discussion about opportunities but also perils and pitfalls of applying automated, computer-assisted text analysis. Several participants also discussed potential avenues for future research.

We thank all the participants, especially the methods experts, for their valuable insights and feedback.

Author: Sören Stapel