Frames in Production: Actors, Networks, Diffusion (FRAMENET)

Lead Research Organisation: University of Warwick
Department Name: Politics and International Studies


Political actors do not present issues objectively. They emphasise certain aspects and deemphasise others and influence the way the audience thinks about the issue, which is called a framing effect. A forest, for example, can be framed as a resource pool to be exploited, a source of artistic inspiration, a fragile and complex ecosystem, or a threat that must be tamed. Each of these alternative frames points to a different policy prescription. Moreover, not every frame is equally influential on its audience. Existing research demonstrates that emotionally compelling frames with negative information are especially effective in changing people's minds. In the recent decades, in most referendums relating to the European Union (EU), emotional arguments highlighting the risks of an increase in immigration played a significant role in persuading a segment of the public to vote against the EU treaty at hand. The importance of frames is evident in today's world. The ways actors frame issues are shown to matter in the fields of elections, immigration policy, environmental politics, trade negotiations, global health, transparency reforms and more.

Although frames have been studied extensively in fields such as political psychology, social movements, international relations or political communication, the main focus of the existing research is their persuasiveness, in other words the factors that affect their persuasiveness. This project asks a neglected question: Where do frames come from in the first place? Why do actors choose the specific frames they use? The project thus aims to create a new, comparative research agenda that investigates when and how specific statements emerge in a political debate, by which kinds of actors they are proposed, and whether and how they diffuse to others. The project studies frames in five issue areas (trade, immigration, environment, global health, and transparency), with two carefully selected political debates in each of these issue areas.

The project takes four steps in order to achieve its aim. In a first step, it uses a new methodological tool called Discourse Network Analysis (studying the content of arguments together with the networks of actors), in order to trace the emergence
of specific frames in a number of selected political debates, the most important actors involved in the process, and the diffusion networks involved. In a second step, it conducts interviews with these key actors involved in framing in order to investigate the most important factors determining their framing choices and whether their respective institutions have an impact on these choices. In a third step, the project studies whether and how these patterns vary from one issue area to another (trade, immigration, environment, global health, and transparency). As such, the project offers a comparative and
comprehensive answer to a crucial but overlooked theoretical question, with an innovative and mixed-method methodology.

The fourth and final step of the project is to disseminate its findings to academic and non-academic communities, in order to raise awareness on the processes that produce frames in these five key issue areas in a globalising world, encourage policy and media elites to be better aware of how their framing processes and strategies impact the public debates on these issues, and to suggest innovative, inclusive and evidence-based communication strategies. Working in partnership with German Development Institute, one of the leading think tanks for global development policy worldwide, and a range of relevant governmental and non-governmental actors, the project integrates key beneficiaries at every stage of the research. These will be achieved through regular meetings, two policy events in Brussels and Geneva, various policy papers, and finally an analytical tool on our website that will visualise our findings so that policymakers, activists and the general public can easily understand them.


10 25 50
Description The project is still ongoing and it is in data analysis phase. So far, we have uncovered important global networks in policy making in fields such as environment, health, migration, Artificial Intelligence, gun control policy and energy security.
Exploitation Route The project is still ongoing.
Sectors Communities and Social Services/Policy





Democracy and Justice

Security and Diplomacy

Description The project is still active but we are in the process of organising policy events and writing policy briefs with our non-academic partner German Institute of Development and Sustainability (IDOS). We are also working on visualisation of our findings to be used by a general audience.
Sector Communities and Social Services/Policy,Digital/Communication/Information Technologies (including Software),Energy,Environment,Healthcare,Government, Democracy and Justice,Security and Diplomacy
Impact Types Policy & public services

Title Improvement of the software (Discourse Network Analysis) 
Description Policy debates are often structured by competing advocacy coalitions around contested policy beliefs grounded in diverging moral concepts. Discourse network analysis has been used to study such polarized policy debates. However, some policy beliefs actors hold are more polarizing or structure-inducing than others. A. It is important to identify such beliefs, both from a normative perspective to facilitate consensus building among actors and their coalitions and from an analytic perspective to improve the homogeneity of the set of policy beliefs that are coded in the research process. In our project, we suggest methods for tracing the relative contribution of each policy belief or set of policy beliefs to structuring a policy debate. We iteratively subdivide the set of coded policy beliefs into an in-set and an out-set (with at least one belief), measure the difference in modularity and the spectral distance between the full discourse network and the discourse network without the out-set as a fitness score of the out-set, penalize the out-set for size, and use combinatorial optimization to find the optimal set of policy beliefs that are immune to structuring the policy debate into camps by maximally deviating from the structure imposed by the remaining policy beliefs. We apply this method to several empirical policy debates to identify the most and least polarizing policy beliefs and make recommendations for promoting dialogue and improving case-based discourse network analysis by optimizing the codebook and analysis. This contribution speaks to the problem of measuring the centrality of policy beliefs in discourse networks, the relative importance of policy beliefs in a given discourse, and the sources of polarization in policy debates. 
Type Of Material Improvements to research infrastructure 
Year Produced 2022 
Provided To Others? Yes  
Impact It is very early in the project. At the moment, we are training 6 early career scholars and roughly 10 undergraduate and postgraduate researchers on this new methodological approach. 
Title FRAMENET dataset 
Description Collecting and coding more than 30000 press articles from press agencies, we identify 500 distinct ways to perceive policy problems and express solutions, used by more than 400 actors over a period of 20 years (2001-2021). This coding covers five issue areas: Trade, Immigration, Environment, Global health, and Transparency 
Type Of Material Database/Collection of data 
Year Produced 2022 
Provided To Others? No  
Impact Our publications will advance the interdisciplinary field of frame analysis both theoretically and methodologically. Our theoretical pieces will focus on the emergence of specific frames in the selected political debates in the issue areas of trade, immigration, environment, global health, and transparency (using content and network analysis); the most important factors determining the actors' framing choices and whether the institutional context in which they communicate their arguments has an impact on these choices (using interviews); and finally whether and how these patterns vary from one issue area to another in a comparative framework. Our methodological pieces will focus on the methodological advancement of the Discourse Network Analysis method, which is a new and promising tool for analysing actors and networks in policy debates based on text data. Our impact plan culminates in the development of a user-friendly analytical tool to enable international policymakers, lobbyists, activists, and the public to better understand and identify the production processes that influence key debates in today's global politics (case studies focus on trade, immigration, environment, health, and transparency), revealing the power structures behind frames, the mechanisms of post-truth politics, and facilitating evidence-based and inclusive responses to and communication strategies around such concerns. 
Description German Development Institute 
Organisation German Development Institute
Country Germany 
Sector Public 
PI Contribution We have secured the partnership of the German Development Institute (DIE), a leading think tank for global development policy worldwide (42,000 Euros in-kind contribution). The three PIs will co-author at least three policy papers with DIE experts in trade, immigration and the environment (Brandi, Martin-Shields, Richerzagen). DIE will also host our two policy events, in Brussels in Year 2 and in Geneva in Year 3, to which we will invite up to 50 international policymakers, elected officials, activists, media outlets, and the public. DIE will send out invitations via their contact list (38.000 contacts including decisionmakers and NGOs).
Collaborator Contribution DIE has a highly respected reputation for its evidence-based policy advice and its expertise in innovative knowledge diffusion via social media tools, from which the partnership will enable us to benefit. Its network includes the European Commission, G20, as well as international policy networks in many emerging economies, providing a worldwide platform to disseminate our findings. We will be in touch with them monthly throughout the project. Also, we will jointly develop an analytical tool which will be hosted on our project website. The tool will involve data visualisation, and allow users to enter specific keywords to track the emergence and diffusion of frames, across our case study areas, to better understand these complex processes and shape their policy and communication strategies accordingly. We will follow up with our stakeholders regularly and monitor citations in their publications to track the impact our research is having on their activities. Drawing on our stakeholder networks and the expertise within our university impact and communications offices, we will produce press releases and policy briefs to disseminate our key project findings to groups with whom we had not already directly engaged, thus maximising our impact potential.
Impact It is very early in the project.
Start Year 2021