Supporting reflective public thought with ConsiderIt

Travis Kriplean, Computer Science & Engineering, U. Washington
Jonathan Morgan, Human Centered Design & Engineering, U. Washington
Deen Freelon, Communication, American University
Alan Borning, Computer Science & Engineering, U. Washington
Lance Bennett, Political Science, Communication, U. Washington

There are surprisingly few intuitive tools for supporting large groups in making decisions together, whether they are citizens, employees, or even program committee members. This is problematic if we are to address challenges that we face as a society. We have invented a new model of public deliberation and have implemented it as the ConsiderIt platform. By encouraging people to think through tradeoffs together and consider the perspectives of others,  we believe ConsiderIt can help build public trust while improving upon our collective ability to take more effective action on problems such as financial reform and climate change.

Unfortunately, I cannot embed a video here, but I am tricking all of you by posting an image of an embedded video which, when clicked, will take you to the video. Watching the video will probably be better than reading the following text. Start at 2:00 if you want to skip to the screencast. Apologies for the lame clip art.

ConsiderIt focuses people on thinking through the tradeoffs of a proposal, such as a ballot measure in an election or whether an academic paper should be accepted to a conference, by inviting them to create a pro/con list which captures the most important factors in their decision. The twist on a traditional pro/con list is that beyond authoring their own pro and con points, participants can also include into their own lists the points others have already contributed. The pro/con considerations of others thus becomes grist for one’s own personal deliberations. By downplaying the role of direct discussion with others and instead focusing on augmenting personal deliberation, ConsiderIt mitigates the activation of political identity and personal attacks. In addition to crafting a personal pro/con point, users summarize their stance on the ballot measures on a sliding scale from strong support to neutral to strong oppose.

ConsiderIt also provides a novel model for participants to explore what other people think, in the form of an evolving guide to public thought on the respective issues. ConsiderIt surfaces the most salient overall pro/con considerations based on how many pro/con lists they have been included in and whether they are included by people with different stances on the issue. Moreover, participants can drill down into the salient points for different segments of the population: “What were those who strongly opposed this thinking???” Users can thus gain insight into the considerations of people with different perspectives, rather than making assumptions based on caricatures. ConsiderIt’s point ranking metric boosts pros and cons that resonate with both supporters and opposers, helping identify common ground.

LVG bus ad
In 2011, the Living Voters Guide could be seen on billboards and buses. For some reason, this made it much more real of an accomplishment for my parents.

Our flagship application to date is the Living Voters Guide (LVG), a crowd-sourced voters guide for electoral deliberation, deployed for Washington state’s 2010 and 2011 elections (browse the 2011 LVG). The LVG has been, in our humble opinion, a success, attracting ~20k Washington residents from across the state through its first two election seasons. Average time on site is over five minutes, suggesting that once citizens come to the guide, they are engaged by the experience. Our analysis indicates a number of normatively desirable outcomes, such as nearly 50% of all pro/con lists containing both pros AND cons. As we found in a user study, citizens were surprised to see people with whom they disagreed recognizing the importance of both pros and cons, even some of the same points that they themselves thought were important. This highlights that (1) in our current mass media environment, we are so used to hearing strong, uncompromising advocates on either side lash out at each other that we are surprised when we hear someone recognizing tradeoffs; (2) there is hope that we can build a more trusting civic space and political discourse by circumventing mass media and relying on direct communication between citizens.

We are currently working to find an institutional home for the Living Voters Guide in libraries nationwide, starting with a collaboration with the Seattle Public Libraries. Aside from helping scaling the LVG up across the country and sustaining it over time, the libraries will be developing a fact-checking service for user contributed pros and cons.

ConsiderIt has potential application beyond the electoral context. We are particularly interested in civic learning applications such as facilitating classroom deliberation about current events, potentially connecting classrooms from different parts of the country. I have also created an extension of ConsiderIt suitable for academic paper peer review. Look for this in the coming months.

Let us know about your ideas for extending the model and where else it might be applied!

For a longer, A/V introduction to this work, watch my recent Phd defense (ConsiderIt specific material starts at 35:14).
For an archival version of the material, see our full paper.

About the author

Travis Kriplean

View all posts


  • Does seeing these pro/con lists cause people to retrench and more strongly back their original positions, or does it cause them to moderate? There have been some interesting psych studies that demonstrate how easy it is to cause people to retrench.

    • We do not know in a rigorous quantitative way if it does. It would be a great followup.

      What we do know is that there was substantial evidence that people were recognizing the legitimacy of arguments from both sides (including both pros and cons), as well as recognizing arguments made by people they disagreed with. It is, however, unclear if this translated into a stronger or weaker stance on the issue. But even that might not signal entrenchment, so much as making up one’s mind given the arguments.

      In our user study, we also saw hints that people reconsidered the other side’s arguments b/c they could actually see the group of people with whom they disagreed recognizing both pros and cons (i.e. actually being reasonable), even recognizing some of the same points that they agreed with (i.e. common ground). This suggests that exposing a user to a group of people who are recognizing tradeoffs might actually act against entrenchment. All of the studies of entrenchment I have seen have resulted from polarization in a group that focuses on differences rather than common ground, which is something that will typically happen in a group discussion.

  • This is fascinating work and exactly the kind of social computing project I’d like to see more of. Congratulations on your success and thanks for sharing!