Measuring citizen frames and their integration in the international policy process: a hybrid BERTopic modelling approach

Abstract

Governments and international organizations are exploring new ways of engaging citizens in the policy design process, however scalable ways to test whether citizen frames appear in resulting policies documents remain limited. We present a hybrid BERTopic modelling approach that extracts frames from short, open-ended citizens’ responses and evaluates their uptake in policy texts by mapping model topics to the Boydstun Policy Frames taxonomy. We demonstrate how heterogeneity in citizen framing across demographic and socioeconomic groups can be measured, offering policymakers a way to account for these differences when drafting inclusive policy documents. We illustrate the method using the Civil 20 India 2023 International Grassroots Survey (2,846 respondents across 64 countries) and the G20 New Delhi Leaders’ Declaration. Our findings suggest only moderate alignment between citizens perspectives and the policy document. The sharpest divergence concerns citizens’ emphasis on the quality-of-life and morality frames, which are largely absent from the outcomes of policy deliberations. Conversely, the policy text leans toward future-oriented, technology centered solutions, while citizens emphasize immediate essentials such as safety and decent wages. The contribution is a generalizable workflow for measuring citizen frames in short texts and assessing their policy uptake, which can be used as a diagnostic tool for auditing democratic responsiveness in global governance.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 1 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.