Initiative awards five research grants to teams of UW faculty members

Project team
Matthew Powers, Professor of Communication
Patricia Moy, Christy Cressey Professor of Communication
Adrienne Russell, Mary Laird Wood Professor of Communication

Project summary
Passed in the 1960s and 70s, open-records laws in the United States were crafted to enable public oversight of government activities, under the assumption that most users would be journalists. Today, at both federal and state levels, the number of requests has skyrocketed – yet news organizations constitute fewer than five percent of public-records requesters.

To better understand this disconnect, our study examines the nature of public-records requests and their impact on requesters, public agencies, and democratic governance. We pose three specific research questions: Who uses public-records requests? For what ends? And with what consequences for democratic oversight? In order to answer these questions, we will: (1) request and analyze records requests to the Washington State Department of Ecology; (2) conduct interviews with its public-records officers and managers as well as the journalists who made the requests; and (3) analyze news articles based on these requests. Answering these questions is critical to renewing and strengthening open-records laws so they better serve their original goal: enabling journalists and other stakeholders to exercise democratic oversight in the public interest.

Our findings should inform efforts by our community partner, the Washington Coalition of Open Government, that allow systems to better serve journalists as well as advocates and civil-society groups working in the public interest. In doing so, it will contribute to the renovation of a key measure whose aim is to make government agencies more responsive to public oversight and input.

Project team
Marie Spiker, Assistant Professor, Department of Epidemiology
Jennifer Otten, Professor, Department of Environmental & Occupational Health Sciences
Sarah Collier, Assistant Professor, Department of Environmental & Occupational Health Sciences
Jillian Youngblood, Executive Director, Civic Genius

Project summary
Civic health is thwarted by public discourse that tends to highlight polarization and overlook common ground. For example, while many perceive polarization about climate change’s existence, only 11% of Americans are doubtful of climate change. We lack nationally representative data to help us to identify common ground on food topics. Food has the potential to be a great unifier—the idea of breaking bread to strengthen social ties is nearly universal—but food is also increasingly at the heart of political debates and divides. Headlines highlight disagreements over rising food prices, public spending on food assistance, Farm Bill funding, food additive regulation, and more. These areas are complex, and we need nuanced reporting that results in action instead of impasse.

Our long-term vision is for an interdisciplinary research collaboration that combines national surveys, qualitative research, journalist workshops, and deliberative democracy methods to leverage food-related issues as a source of common ground to cultivate greater civic health. In the short term, funds from this Civic Health Award will catalyze this work and increase the likelihood of future funding by enabling us to (1) develop, field, and analyze data from a national survey of US consumers and (2) conduct focus groups with local Washington journalists.

These findings will provide insights on: Where do Americans have common ground on consequential food-related issues? How can an understanding of this common ground guide journalists in more effective communication on complex topics? And how can recent WA legislative investments in local journalism be leveraged for civic health?

Project team
Cory Struthers, Assistant Professor, Evans School of Public Policy & Governance
Esther Min, Director of Research, Front and Centered

Project summary
Environmental and social justice policies often include provisions to strengthen community input in decisions that affect them. Co-governance is a promising approach, wherein communities and government co-create policies and programs. However, there is growing evidence that co-governance may reinforce systemic inequalities, limiting the voices of those who are furthest from economic well-being.

In 2024, the Washington State Legislature allocated $2 million to pilot community assemblies as a form of co-governance, designed to counter common patterns of exclusion. While government is involved, community-based organizations lead and design the process, aiming to (i) develop community-driven recommendations, and (ii) strengthen their long-term civic capacity. Our proposed research will identify the mechanisms through which community assemblies foster these outcomes, examining the eight community assemblies piloted in Washington state. We will use process tracing, combined with systematic content analysis, to trace how participants’ civic capacity and assembly recommendations develop during assemblies. This analysis leverages evidence, including participant surveys, meeting transcripts, notes, and interviews.

Funding from the Civic Health Initiative would support analysis and data collection, including interviews to reinforce inferences from process tracing. The community assembly model is timely, as scholars and practitioners reckon with exclusion challenges in other forms of co-governance. Washington state’s pilot assemblies are a unique opportunity to examine their efficacy, and understand the mechanisms through which they meet their goals. Findings will be used to develop a theory change that informs future research and evaluation on co-governance, including community assemblies in and beyond Washington state.

Project team
Amy Zhang, Assistant Professor, Allen School of Computer Science & Engineering
Ruotong Wang, PhD student, Allen School of Computer Science & Engineering
Sam Wong, Global Innovation Exchange

Project summary
Online civic discussions are important for shaping local policies, yet participation remains skewed towards dominant and expert voices. Politically disengaged individuals often hesitate to contribute due to their perception of the lack of expertise in the issue. While prior scaffolding approaches like structured pros and cons or ideation prompts have shown promise in encouraging participation, they often lack personalization and scalability.

This project explores how large language models can enable personalized scaffolding at scale to support more inclusive civic discourse. We propose an online discussion platform that integrates two key components: (1) Interactive Summaries to help users quickly make sense of ongoing conversations, and (2) Personalized Reflection Nudges that helps users reflect on the existing discussion and themes. These nudges are tailored to the user’s interaction with the discussion, their comment drafts, and their viewing history. We will create novel LLM workflows by leveraging techniques such as retrieval-augmented generation (RAG) to generate these summaries and nudges.

To evaluate how these features impact the comment writing process, we will conduct both a controlled lab study and a real-world deployment in collaboration with civic deliberation platforms such as ConsiderIt. In both studies, we will assess the effectiveness of our approach by rating the (1) Comment Quality such as content relevance to the discussion and references to earlier comments, (2) Reflection capacity measured with the Self-Reflection and Insight Scale (SRIS) and (3) User satisfaction with the nudges. This work contributes new methods for lowering participation barriers in civic discourse and offers practical design insights for building inclusive, LLM-supported discussion platforms.

Project team
Ying Yang, Assistant Professor, UW Bothell School of Business

Project summary
This 12-month pilot project investigates how AI-augmented peer feedback systems can motivate Generation Z students to develop civic knowledge, trust-building skills, and collaborative mindsets—key attributes of civic-minded leadership. While Gen Z students care about social issues, many are disengaged from traditional civic education due to perceived lack of relevance or motivation. This study integrates behavioral economics and educational research to test whether structured peer evaluation— augmented by AI feedback—can enhance student engagement and civic learning.

The project will be embedded in a university course where students submit work on civic or policy topics, evaluate peer submissions, and receive feedback from both peers and an AI tool. The intervention compares student outcomes across sections with and without AI-generated prompts, testing whether augmented feedback increases critical thinking, civic identity, and perceived relevance of civic issues. The approach promotes “learning by evaluating,” where students gain insight through evaluating others’ work—a model supported by cognitive learning theory.

Data collection will include surveys, quizzes, and analysis of student feedback to assess learning gains, feedback quality, and changes in civic attitudes. The pilot will demonstrate proof-of-concept for a scalable civic-learning model and support future proposals for longitudinal studies or broader curricular integration across the UW system. The project aligns with the Civic Health Initiative’s mission to prepare civic-minded leaders and scale inclusive, accessible civic education.