During the reporting period, progress advanced across all four work packages, with emphasis on data collection, dataset preparation, and alignment with the finalized theoretical framework. The project’s key theoretical contribution is the extension of cleavage theory to the global level. By adapting the exit-voice framework to system building, a typology was developed distinguishing four forms of politicization of global inequalities—universal, functional, territorial, and territorialized. This framework, published in a peer-reviewed output, guides ongoing empirical work, coding schemes, and methodological choices. Two GLOBAL Working Papers are in preparation, and all work packages have been aligned to the theory.
In Work Package 1, data collection faced challenges due to fragmented sources and the specificity of one-party systems in non-Western contexts. A “party system tree” method was devised to address these. Task 1.2 shifted to applying large language models to comparative literature, enabling systematic coding of cleavage structures in 12 polities (completion expected in 2025). In Task 1.3 AI-assisted instruments were designed to map transnational networks across organizational, financial, military, and personal linkages, with coding underway.
In Work Package 2, data collection for Tasks 2.1–2.3 is nearly complete, while Task 2.4 faced issues of availability and cost of newspapers, partly resolved with support from the host institution’s library. A key achievement is the development of a validated large-language-model classification system applying the theoretical typology to diverse corpora, including manifestos, institutional texts, and press sources. The method, already presented in a conference paper, proved robust and particularly effective for Catholic Church texts.
In Work Package 3, the main dataset of UN General Assembly roll-call votes was completed, producing ideal-point estimations that capture global alliances and cleavages. Additional socio-economic, demographic, and organizational variables are being added, while General Debate speeches provide an alternative measure of ideological positioning. Task 3.1 was redirected toward two lines of work: classification of over a century of national legislation across 12 polities, and a nearly finalized database of international treaties. These resources will enable analyses of regional vs global politicization in line with the theoretical framework.
In Work Package 4, the design of a cross-national survey in 12 countries was finalized. It introduces innovative instruments to measure identity, solidarity, and political action, including a 16-item battery reflecting the four cleavage types and a conjoint experiment. The survey has been validated by experts, approved by the ethics committee, and contracted to YouGov for fielding in Fall 2025, with translations into major local languages already prepared.
Overall, the reporting period consolidated the theoretical foundations of the project, introduced novel data collection and classification tools, advanced multiple datasets, and prepared a global survey, laying a strong basis for the next stages of analysis.
