In person icon Innovative Computational Methods for Interest Group Research

Abstract

Quantitative text analysis has been widely employed to study large bodies of documents in the interest group literature. Researchers have notably used quantitative scaling methods, such as Wordfish, to identify a latent ideological dimension based on word frequency patterns. These approaches have been particularly influential in analyzing interest groups’ consultation submissions. However, they have also faced criticism for being ill-suited to the characteristics of these texts. A key challenge lies in whether consultation documents can be meaningfully aligned along a single latent policy dimension, as assumed by text scaling methods. Furthermore, word frequency-based approaches struggle to capture policy stances in highly technical or bureaucratic language, where nuanced meaning is conveyed through phrasing rather than simple word usage. As a result, scholars studying interest group influence and success have often shied away from purely quantitative text analysis, instead relying on qualitative assessments or manual coding. Still, we believe that computational text analysis holds great potential for advancing the study of interest groups, particularly as new methods move beyond traditional text scaling. In this panel, we seek contributions that explore alternative computational approaches, including: • Word embeddings to measure text similarity based on word co-occurrence rather than raw frequency. • Machine learning-based parsing to systematically extract and attribute statements to interest groups in media coverage, allowing for a more fine-grained analysis of their role in agenda-setting. • Structural Topic Modeling (STM) to uncover variation in framing strategies across different types of interest groups and policy areas. • BERT-based classification models to automatically detect policy positions or argumentation styles, improving our ability to analyze how interest groups communicate their demands. • Network analysis of co-occurring actors and themes, identifying alliances and coalitions among interest groups within public debates. By incorporating these approaches, we can gain a more nuanced understanding of how interest groups position themselves, interact with policymakers, and shape public discourse. This panel invites contributions that critically assess, refine, or apply such methods to advance the study of interest groups in contemporary politics.

Title	Details
Geopolitics Meets Economic Interests: Business Lobbying Amid Security Concerns	View Paper Details
How to Measure Interest Group Voice in the News Media? Addressing the Unitization Challenge	View Paper Details
Who Speaks Like Citizens? Investigating Citizen-Interest Group Congruence in Stakeholder Feedback to European Commission Consultations	View Paper Details
Digital Media Detection System for Interest Groups in the European Union	View Paper Details
The Politicization of Belt and Road Initiative (BRI) Media Coverage: A Computational Text Analysis	View Paper Details