Being female, liberal, intellectually humble, and having weak party identification are all positively associated with writing more persuasive political arguments, according to new research published in PNAS Nexus. The study provides insights into the factors that influence perceived persuasiveness.
The researchers were motivated by the importance of political persuasion in a democratic society, where citizens engage in sincere attempts to sway the opinions of others. They believe that understanding the attributes of persuaders that contribute to their effectiveness can advance knowledge on how interpersonal persuasion can facilitate political tolerance.
“I wanted to study how average people attempt to persuade one another about politics. Scientific work on political persuasion tends to focus on top-down persuasion: political advertisements, politicians’ speeches, etc,” explained study author Jeffrey Lees, a PhD associate research scholar at Princeton University.
“Yet a healthy democratic culture requires regular people who disagree with each other to converse, debate in a civil manner, and even change their minds. I wanted to examine what people would say if given the freedom to write about any political topic they wanted, and how persuasive their arguments would be.”
The researchers aimed to identify the individual characteristics that predict perceived persuasion for both in-party and out-party communication, while controlling for confounding factors. To conduct the study, they recruited two groups of participants: persuaders and judges.
The persuader sample consisted of 605 Democrats and Republicans, evenly split based on a prescreen, who were recruited on the Prolific survey platform. The participants were asked to write a persuasive political argument of at least four sentences on any political topic they wished.
The prompt varied for different participants, with some assigned to persuade an in-party member, some assigned to persuade an out-party member, and others assigned to persuade the “average American.” The participants were informed that their arguments would be read and rated by members of the group they were assigned to persuade.
The persuaders also provided self-ratings of their own arguments and opinions, as well as meta-ratings of how they believed the “average” member of their target group would perceive them and their arguments. The persuaders’ demographic information was collected, and they were compensated for their participation.
The judge sample included 3,131 participants (1,574 Democrats, 1,293 Republicans, and 263 Independents) from the United States. Participants were recruited through a Forthright Access survey panel, and efforts were made to match the sample’s demographics to the US census.
The judges were informed that they would be reading six political arguments from persuader participants. They were randomly assigned to read four arguments written for their party (Democrats or Republicans) and two arguments written for the “average American.” Independents read four arguments written for the “average American” only.
Each judge rated the arguments they read on various measures, which were similar to the self-ratings and meta-ratings provided by the persuaders. The judges also provided demographic information and received compensation for their participation.
The results showed that condition strongly affected persuasiveness, with arguments targeting in-party members perceived as more persuasive and arguments targeting out-party members perceived as less persuasive. Perceived persuasiveness was also found to be positively associated with intellectual humility and argument length, while negatively associated with party identification strength.
“We found that people across the aisles are willing to listen to arguments from people on the ‘other side,'” Lees told PsyPost. “And in many cases, they rated those arguments as highly persuasive. So a central takeaway is that we’re not all as polarized as we tend to believe, and if you make a sincere attempt to respectively express your political opinions to someone who may disagree with you, they’ll likely listen.”
“Also, we found that people who were more open to the idea they might be wrong, a trait we in psychology call intellectual humility, were more persuasive. This suggests that when attempting to persuade someone to your side, claiming to have absolute certainty about your position can actually hurt you. Instead, acknowledge that everyone is fallible, and that you too are willing to change your beliefs in the face of new information.”
Gender was consistently associated with perceived persuasiveness, with women’s arguments being rated as more persuasive than men’s. This gender difference persisted even when controlling for demographic factors and other variables. Moreover, gender did not interact with the condition, indicating that women’s arguments were rated as more persuasive across all targets of persuasion.
The researchers also conducted sentiment analyses to explore why women and liberals were perceived as more persuasive. However, adding sentiment scores to the model did not fully explain the gender and ideology effects.
“We were really surprised to find that women were more persuasive, on average, then men,” Lees said. “This seems to be driven in part by two factors. First, women put more effort into writing their arguments, which were significantly longer than men’s, and longer arguments were rated as more persuasive.”
“Second, men used more words expressing dominance in their arguments than women, which made them less persuasive. Yet these factors didn’t entirely explain women’s superior persuasiveness. We’re hoping that future research might reveal why women write more persuasive arguments.”
Additionally, the study analyzed the presence of political topics in the arguments and found that certain topics, such as economic inequality and healthcare, were rated as more persuasive than others.
But the study, like all research, includes some caveats.
“An intentional decision we made was to conceal most information about who wrote the arguments from the people reading them,” Lees told PsyPost. “So, for example, the people reading the arguments and rating how persuasive they are did not know the gender of the person who wrote it.”
“So an important caveat is that our findings are most reflective of written persuasion, but may be less reflective of spoken, face-to-face persuasion. For example, sexist attitudes toward women might lead people to discount their arguments. This is exactly why we made this design choice, but it does limit the real-world contexts to which our results apply.”
The study, “Women, the intellectually humble, and liberals write more persuasive political arguments“, was authored by Jeffrey Lees, Haley Todd, and Maxwell Barrantie.
