Among the many causes that could explain the defeat of Kamala Harris in the elections in the United States, the excessive emphasis placed in her campaign on the issue of abortion, to the detriment of other issues perhaps more important to many voters. This is a plausible reason. But why was she or her campaign team wrong? Was it a calculation error? An article published this year in the American Journal of Political Science with my co-authors Nikolas Schöll and Gaël le Mens (titled “How Politicians Learn from Citizens’ Feedback: The Case of Gender on Twitter”) suggests that the reason could be a mixture of the normal functioning of campaigns as well as the differential treatment enjoyed by men and women in politics.
Politicians operate in an environment characterized by great uncertainty about the issues that concern citizens. Even if they try to use surveys or other tools, the volumes of information they process are enormous, and a common heuristic (a cognitive shortcut for making decisions) is the feedback they receive from citizens . “Reinforcement learning” is useful because it allows us to make decisions quickly, in this case on which issues to prioritize in a campaign.
However, this mechanism can also introduce bias. In this article, we argue and show that female politicians receive disproportionately higher positive responses when discussing gender issues compared to their male counterparts. Our data concerns Spain, but the argument is general. For Harris, an initially broad campaign platform likely included issues appealing to diverse segments of voters. However, as your campaign has evolved, you may have seen stronger responses on issues such as abortion.
In this article, we explore the mechanisms that explain why female politicians might be more interested in “women’s issues.” The first is intrinsic motivation, meaning that women in politics may be more interested in gender issues. The second is public self-selection, if people more interested in gender follow female politicians more. And the third mechanism is differential treatment based on the sex of the candidate. Maybe we treat men and women differently. We only found evidence for the third mechanism of differential processing.
Although we would need more detailed data to verify whether this conjecture is true, in her concession speech, Harris’s immediate reference to abortion as a key issue was met with enthusiastic applause. On the other hand, he spoke little about other subjects.
Placing too much emphasis on abortion may seem like a misstep, and perhaps it is. Regardless, the article’s analysis suggests that this decision may reflect the normal dynamics of an election campaign as well as underlying biases in the way citizens interact with female politicians, rather than a conscious error in campaign strategy.
