Asymmetries of Vulnerability: How Party Lines Shape Racialized Online Violence Against Women in the 2024 U.S. Election
This study analyzes over 700,000 tweets from the 2024 U.S. congressional elections to demonstrate that partisan identity intersects with race and gender to create asymmetric vulnerabilities to online violence, specifically targeting Democratic women of color and white Republican women with elevated levels of abuse compared to their counterparts.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the digital town squares of modern life, social media promised to give everyone a voice, leveling the playing field for those who had been shut out of traditional political conversations. For decades, researchers have watched how this new public square actually functions, and the picture that has emerged is far less democratic than hoped. Instead of a free exchange of ideas, these platforms have become breeding grounds for harassment, where women in politics are disproportionately targeted with abuse, threats, and insults. This violence is not random; it follows a pattern where a person's identity shapes how they are treated. Scholars have long understood that being a woman makes a candidate a target, and that being a person of color adds another layer of risk. But a crucial piece of the puzzle has remained missing: how does a candidate's political party change the way their race and gender are perceived by an angry online crowd? In a country deeply divided by political affiliation, the meaning of a woman's identity shifts depending on whether she stands with the Democrats or the Republicans. Understanding this shift is vital because it reveals who is most at risk of being silenced, and it challenges the simple idea that all women face the same kind of online danger.
A team of researchers set out to map these shifting dangers during the 2024 U.S. congressional elections. They did not rely on anecdotes or a few viral posts; instead, they built a massive dataset by collecting more than 750,000 tweets that mentioned candidates running for the House of Representatives and the Senate. To make sense of this ocean of text, they used advanced computer programs trained to recognize the difference between a rude comment and a hateful attack. These tools were taught to spot specific types of abuse, ranging from general insults and mockery to speech that targets a person's race, religion, or gender. By linking every tweet back to the specific candidate it mentioned, the researchers could see exactly which groups of people were being targeted and how often. They looked at nearly 1,000 candidates, breaking them down by gender, race, and party affiliation, to see if the pattern of abuse changed when these factors were mixed together.
The results revealed a striking asymmetry in who gets attacked. The study found that the idea of a single "women's vote" or a uniform experience of harassment for all female candidates is incorrect. Instead, the danger a woman faces depends entirely on the combination of her race and her political party. For white women running as Republicans, the data showed a sharp increase in abuse compared to their male colleagues. These candidates faced a predicted probability of receiving offensive speech that was about 4.2 percentage points higher than that of white Republican men. They were also significantly more likely to receive hate speech. This suggests that white Republican women are targeted because they are seen as symbols of conservative gender politics and traditional values, making them focal points for critics on the other side of the political divide.
In a complete reversal, the pattern flipped for women of color. When these candidates ran as Democrats, they faced a much higher level of hostility than their male counterparts. The study found that non-white Democratic women were about 2.5 percentage points more likely to receive offensive speech than non-white Democratic men. This vulnerability was not spread evenly across all groups; it was most intense for Black and Asian women. These candidates faced the highest predicted rates of both general abuse and specific hate speech targeting their race and gender. The researchers suggest this is because these women are viewed as powerful symbols of progressive change and racial inclusion, making them high-value targets for those who feel threatened by shifting demographics and cultural norms.
The study also clarified what does not happen. It ruled out the idea that all women of color face the same level of danger regardless of their party. Republican women of color did not experience the same surge in abuse as their Democratic counterparts; in fact, the data showed that their risk was often lower than that of white Republican women. Similarly, the research showed that white Democratic women did not face the same intense, race-fueled backlash as women of color in the same party. The violence was not a simple matter of being a woman or being a person of color; it was the specific intersection of those identities with a political party that determined the risk.
These findings paint a complex picture of the digital political landscape. The abuse is not a uniform storm hitting everyone equally; it is a targeted force that strikes different groups for different reasons. White Republican women are attacked for what they represent in conservative politics, while Black and Asian Democratic women are attacked for what they represent in progressive politics. The study concludes that to truly understand online violence, we must stop looking at gender, race, and party as separate factors. Instead, we must see how they combine to create unique vulnerabilities. This understanding is essential for anyone trying to protect the integrity of democratic participation, as it shows that the path to the ballot box is made significantly more dangerous for some women than for others, depending entirely on the political banner they carry.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.