Public engagement and deliberation play key roles in democratic society. Yet, civic engagement is uneven at best, and thoughtful public deliberation about major issues is often displaced either by apathy or shrill and extreme voices. This project develops new ways for citizens and government to communicate using social-computational systems. The project develops systems that facilitate adding crisp, relatively neutral summaries alongside advocacy statements; jointly-authored position statements, with flexible ways to sign on to existing statements, fork new ones, and understand and track changes; and enhanced moderation techniques. A partnership with the City of Seattle is enabling the researchers to test their ideas and systems in actual use.
The research studies a set of difficult and unsolved research issues in social-computational systems. How can systems support alternate divisions of labor among citizens and government, including effective support for automating some tasks and making the remaining tasks easy for human participants to work on? How can systems effectively support understanding a range of positions, and highlight who is agreeing or disagreeing and why? How can engagement systems allow and encourage effectively building on others' ideas and opinions? How can they handle strategic activity in which individuals or groups flood the system with duplicate posts, or create multiple accounts to give the impression of widespread support? How can input be effectively summarized and presented? The broader impacts of the work are substantial: if successful, this research will strengthen democratic process by facilitating effective citizen participation and deliberation.
Our goal in this project was to develop socio-technical systems to support civic deliberation and participation, in partnership with community groups. We sought to develop systems that help participants listen to others (particularly those with different points of view), consider tradeoffs, and seek common ground. An important aspect of the work has been to deploy these systems for real issues and elections, and evaluate them in real use. Regarding outcomes for the work as listed in the original proposal, in the first year we proposed to undertake a review of related systems, identification of key conceptual issues that should be part of the design and implementation plan, development of initial system prototypes, and launch of initial version of application. This has been accomplished. The most successful project is the Living Voters Guide, a crowd sourced voter deliberation platform for Washington State initiative elections. We proposed that system evolution would continue in years 2 and 3, making design improvements and expanding the user base. These goals have been accomplished as well. Among the related outcomes from this successful deployment of Living Voters Guide are the following: Living Voters Guide Deliberation Platform By most standards Living Voters Guide is a successful exemplar of civic technology implementation. LVG successfully expanded its diffusion from the city of Seattle to the state of Washington within the scope of 3 years, attracting around 20,000 unique visitors. (See images for views of this platform.) A version of the ConsiderIt software that is the basis for the Living Voters Guide is under an open source license and is freely available via the github repository. Travis Kriplean, who completed his Ph.D. dissertation while supported by this grant and who was subsequently a postdoc on it, co-founded a small company to support deliberative technologies like the Living Voters Guide. This activity has been undertaken in collaboration with and with the approval of the University of Washington Center for Commercialization, and the Commercialization Oversight Committee from the UW Department of Computer Science & Engineering. This activity will provide another major venue for dissemination and use of the results. Training and Development The graduate students employed on this project at the University of Washington have gained valuable skills in research instrument design as well as data transformation, analysis, and visualization. These students included: Travis Kriplean and Caitlin Bonnar from Computer Science & Engineering; Jonathan Morgan from Human Centered Design and Engineering; Deen Freelon, Sheetal Agarwal and Coutrney Johnson from the Communication Department; and Caterina Rost from Political Science. Three dissertations (Kriplean, Freelon, and Agarwal) have been supported by, and used data from this project. Further, there has been much fruitful cross-disciplinary exchange, with the HCI experts learning much about the practice of social science and vice versa. Best Practices for Civic Technology Development We have also been able to distill lessons from the Living Voters Guide to yield "best practices" for civic technology. These results will be presented in more detail in Sheetal Agarwal’s forthcoming Ph.D. dissertation. We summarize them briefly here. Governmental institutions, though often identified as the primary candidates for civic technology projects, are not always the best-fit candidates due in part to bureaucratic processes. Innovation requires a certain flexibility and fluidity that institutionalized rules and regulations can limit and delay as developer teams navigate through red tape. Forging partnerships with civic organizations, such as CityClub of Seattle, loosens bureaucratic restrictions and creates opportunities to move from creative ideation to design and development in an efficient and timely manner. Other civic institutions such as the public library have also proved valuable and receptive partners. For example, a fact-checking service that became part of the LVG was a successful collaboration among the University of Washington, Seattle CityClub, and the Seattle Public Library. We found that users demonstrated demand for fact-checks and the willingness to submit requests for them. Dissemination of Results to Communities of Interest A number of publications have issued from this project, including articles placed in leading journals, papers in conference proceedings, and a book by a top university press. These publications are listed in the annual reports on the project. These publications have generated interdisciplinary attention in computer science, political science and communication. Key materials have also been circulated to our partners at Seattle CityClub and the Seattle Public Library, who found it quite useful in understanding the results of the Living Voters Guide. Improving Civic Engagement Civic engagement and an informed electorate are cornerstones of a healthy democracy. We believe that, in a small way at least, the deliberative technologies developed under this grant can help support both engagement and developing a better-informed electorate.
