Converj’s deliberative forms resemble a traditional survey with a simple sequence of questions. However, Converj also incorporates lightweight structured deliberation, where participants can share proposals with pro’s and con’s, and read other participants’ answers and reasons.
Problems and Purpose
Most social-networks separate us into filter bubbles of like-minded people. These groups are comfortable, and the more time we spend there, the more we hear our own opinions reflected back to ourselves, and the more extreme we become in our views. When we leave our comfortable bubbles to make controversial decisions for the larger society, our conflicting extreme opinions create chaos. As people move further away from the mainstream, decision making for our city, state, and nation becomes more difficult. We need a way to understand the world outside our own bubble, to encounter diverse information and values. We need a way to re-converge our isolated islands of belief, back into a single mainland society. We need a way for people to disagree, yet still understand each other, and understand the facts. We need a way to plan a future together.
Deliberation bursts filter bubbles, nudging participants to think about their own reasoning, and to be aware of others' perspectives. Deliberation decreases polarization and establishes common ground. This opening up encourages consensus in diverse groups, and improves the quality of group decisions. Yet many leaders hesitate to use deliberative forums, because of the risk that the discussion may go off topic, may devolve into insults, or may be too complicated.
Converj’s deliberative forms ease the transition from traditional opinion survey forms to deliberative forums. Converj forms are set up like a traditional survey, with a simple sequence of questions. But Converj forms also incorporate a lightweight type of deliberation, to improve the quality of participant feedback.
Origins and Development
Converj forms are offered completely free, from a privately held company which seeks profit only to sustain operations. Converj is led by Chad Brower.
Converj has independently arrived at solutions similar to ConsiderIt, Tricider, and Kialo, which all support forms of online structured deliberation.
Participant Recruitment and Selection
Once a host creates a survey, they distribute the survey's private link through their group's existing communication systems, such as email lists or group chats. Anybody with the link can immediately participate in the survey. Hosts and participants do not have to register. To protect privacy and reduce groupthink, Converj does not track users' identity nor reputation, letting each answer stand on its own merit.
How it Works: Process, Interaction, and Decision-Making
Converj’s deliberative forms ease the transition from traditional opinion survey forms to deliberative forums. Like a traditional survey, Converj forms can be set up with a simple sequence of questions, including rating, ranking, checklist, multiple choice, write-in, and budget questions. Hosts can suggest answers, and participants can also add new answers.
Converj forms incorporate a lightweight type of deliberation. When a participant answers a question, they are asked to include a reason with their answer, encouraging them to both think more deeply about their preference. To reduce the effort of writing a reason for each answer, once the user starts typing a reason, Converj suggests similar existing reasons which the user can click to copy. Converj chooses which reasons to suggest based how much the user’s reason in progress matches the words in the suggested reason, and based on how many other users voted for the suggested reason. The reason suggestions help the participant quickly encounter relevant, high quality reasons, without browsing through low quality reasons. The reason suggestions nudge participants to consider other people's reasons. And the reason suggestions reduce the number of unique reasons that the host has to review.
In addition to traditional question types, Converj forms support questions that are more deliberative in intent. For instance, Converj forms can include a proposal for participants to vote for or against. In order to vote on a proposal, the participant must provide their best reason for or against the proposal – or select an existing reason from other participants. This nudges participants to reflect and learn, leading to better quality decisions, and leading to more support for the outcome. For participation earlier in the planning process, Converj provides a question type to request proposals about how to solve a problem, with proposals ranked by their pro and con reasons. And even further upstream, for leaders trying to discover the concerns of a community, there is a question type to request problems that need solving.
Little moderation is required, because only the best quality, most relevant user answers are displayed. Immediate results are available to all participants.
With Converj, there is no need to choose between surveys versus deliberation, because deliberation is built into a simple survey form.
Influence, Outcomes, and Effects
Converj automated deliberation improves the quality of group decisions, makes the leader's job easier, and increases participant support for the outcome.
Voting alone does not bring opinions closer together. Competitive voting can even increase partisanship. But Converj's reason-based voting causes people to learn more about their own ideas, and about opposing ideas, finding new common ground. And Converj’s reason suggestions make it easy to express and discover reasons, to make the most of participants' time and effort.
Likewise, Converj makes the host's job easier in several ways. Allowing participants to write in new answers, Converj enables the discussion leader to discover opinions that they did not anticipate. This helps prevent upsetting surprises later in the planning process, and sometimes reveals new options that are superior to the ideas that leaders previously considered. And participants are nudged to discover and vote for a few main answers, to make results both representative and clear for reporting.
With Converj, leaders don't just collect participants' answers, they reveal why people gave those answers, not in terms of demographics, but in participants' own words. Extremist participants are allowed to speak their mind, without disrupting coherent deliberation. Participants can vote for only one top reason, which makes answer and vote spamming ineffective. Answer length penalties reduce the visibility of answers that ramble on. Vote counts are public, yet deemphasized to reduce bandwagon effects. Opposing opinions are given equal placement, side by side, so that minority viewpoints are not drowned out by the majority.
Converj's lightweight deliberation causes groups to provide better quality feedback, with less effort.
Analysis and Lessons Learned
Using Converj, an experiment compared the effect of requiring participants to provide only answers, versus requiring participants to also provide reasons for their answers. The experiment results showed that requiring reasons increased the group's accuracy on objective questions, and increased the group's empathy on questions about subjective values.