A 200-citizen bipartisan debate reveals people's true attitudes towards abortion.
In the politically charged atmosphere surrounding abortion rights in the USA, PSi hosted a groundbreaking dialogue to explore citizens' attitudes towards abortion access in the United States. This case study outlines how PSi facilitated a 200-person live debate among a bipartisan group of people living in America.
The discourse on abortion rights in the USA is profoundly divisive and has further polarised following the US Supreme Court's decision to overturn Roe v Wade, the national law that protected abortion rights. Yet, in our citizen debate, we found two surprising results:
While disagreeing on several points, both Republicans and Democrats expressed views on abortion rights that were considerably more balanced than the vitriolic rhetoric commonly observed online and in the media.
In our discussion, highly polarising ideas did not go far. Polarising ideas, namely ideas that appealed predominantly to one political faction—Republican or Democrat—but not the other, received fewer votes and were discussed the least.
Background
Scholars argue that citizen deliberation can mitigate polarisation, enhance public awareness, safeguard against disinformation, and promote consensus-building. However, managing conversations with more than 10-20 individuals can be challenging. Notable instances of such large-scale citizen deliberations include Emanuel Macron’s Citizens Convention for Climate (2019-2020) and the Irish Citizens' Assembly (2016-present). Unfortunately, logistical and financial barriers render these initiatives unique rather than the norm.
With a commitment to diminish polarisation and foster constructive public discourse, we convened a debate on the divisive issue of abortion rights in the United States. PSi allowed hundreds of simultaneous speakers to engage in a single online debate.
Objective
The event sought to foster a bipartisan dialogue on abortion laws in the United States, inviting 600 participants split evenly between Democrats and Republicans to deliberate on the question: “What do you think the laws around abortion access should be in the United States?”
Methodology
A total of 270 participants proposed ideas, with 208 engaging in the conversation over four discussion rounds.
Participants were randomly assigned to discussion groups, ensuring a mix of individuals who either aligned with their political stance or represented the opposing viewpoint. At the end of each round, participants could distribute a small number of votes to the ideas they found most compelling, excluding their own contributions. They could concentrate their votes on a single idea or distribute them across multiple suggestions according to their preference.
As the debate progressed, the discussion consolidated around the ideas that received the most support in previous rounds.
Results
Feedback from participants was overwhelmingly positive. Many people appreciated the active engagement in the discussion and expressed a willingness to recommend the platform to others.
The entire discussion lasted for only 45 minutes. Yet, leveraging its automated moderation capabilities, PSi was able to collect over 15 hours' worth of dialogue data within this short span. Some of the discussions were enlightening.
The laws in the United States surrounding abortion access need to be dependent upon the situation in consensual sex. The mother has already made a choice, and therefore, we have to concentrate on the child's needs.
Debating is more engaging than surveys.
People were highly engaged in the debate. An engagement score of 72% was observed, indicating active participation. Participants reported that the conversation was highly active (average rating: 4.3/5).
Now, compare this to most public attitude surveys. According to Pew Research, it has become increasingly difficult to contact potential respondents and persuade them to participate. The response rate of a typical telephone survey was 36% in 1997 and is just 9% today.
These results indicate that live online debates can increase engagement, improve access to public attitudes and, as we show below, foster de-polarising bipartisan debate.
A wide range of ideas.
As measured by their semantic similarity, the proposed ideas showed a 62% diversity score, signifying a broad range of perspectives. Perspectives ranged from highly liberal ("Abortion should be legal for anyone who wants an abortion in the United States. And it is a woman's choice.") to highly conservative ("I believe that most abortion should be illegal"). The average correlation among people's votes within a table was r=-0.10, reflecting the fact that our sample was balanced across opposing camps of the political spectrum.
I believe that the law in the United States should be that this is a medical issue that should be left up to the doctor and the person it affects.
Two themes emerged from the discussion.
PSi automatically performs linguistic analyses and semantic similarity scoring to surface insights and common themes. As shown in the idea map below, two main themes were found. The first idea cluster (the blue cluster) was related to the legal aspects of abortion, such as when it should be permitted, until what trimester and who should be responsible for the decision. The second cluster (in red) grouped ideas related to a woman's individual rights.
Because so many abortions were happening, my parents were not allowed to adopt more children. I've seen the other side of this issue.
PSi also automatically estimates relationships between ideas based on people’s voting patterns and visualises the network of ideas' relationships. In this discussion, “Abortion is a woman’s choice” and “Doctors and Patients have exclusive rights to medical decisions” were related to each other and shared a considerable pool of supporters. In contrast, the idea that “Laws surrounding abortion access need to consider the cause of the pregnancy as well as the needs of both mother and child.” stood by itself among the top-voted ideas.
Notice how, despite originating from participants of contrasting political leanings, the majority of ideas proposed expressed more balanced views than the often vitriolic rhetoric prevalent in media and online discussions about abortion.
Less polarising ideas go farther.
The discussion revealed a striking pattern: the level of an idea's polarisation inversely correlated with the number of votes it received. Polarisation was determined by an idea's appeal predominantly to one political faction—Republican or Democrat—but not the other. Conversely, ideas that resonated more evenly across the political divide were deemed non-polarising.
We find that less polarising, more moderate ideas that appealed to both Republicans and Democrats were deliberated upon more extensively and garnered increased support throughout the rounds. This trend underscores PSi's potential to mitigate polarisation and encourage bipartisan discourse, even on the most contentious issues.
Conclusion
Abortion rights will be a pivotal issue in the 2024 presidential elections. The PSi platform demonstrated its capacity to host substantive and balanced discussions on contentious topics at an unprecedented scale. The case of abortion rights in the USA showcased PSi’s ability to not only handle the complexities of large-scale deliberation but also to extract meaningful, action-oriented insights. We will continue to show the power of collective intelligence to foster understanding and reduce polarisation.
Check the discussion's data dashboard here.
About PSi
PSi has innovated a technique capable of scaling live voice dialogues to encompass hundreds or thousands of participants concurrently. The platform harnesses collective intelligence principles to counteract social biases such as polarisation, herding, groupthink, and the disproportionate impact of dominant voices.
If you want to use PSi with your stakeholders, get in touch.
Get in touch
