Running While Female: Using AI to Track how Twitter Commentary Disadvantages Women in the 2020 U.S. Primaries

Oates, Sarah; Gurevich, Olya; Walker, Christopher J; Meco, Lucina Di

doi:10.2139/ssrn.3444200

Cited by 8 publications

(15 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We combined a top-down approach that relied on prior work of similar topics with a piloting and bottom-up testing period. In the initial definition of our codebook categories, we relied on prior research on the various topics that were a critical part of our study: content analyses and other forms of research about gender (Gordon et al, 2017;Heldman et al, 2018) and race stereotypes (Dovidio et al, 1986;Gaertner & McLaughlin, 1983;Tukachinsky et al, 2017); research on other identity stereotypes such as disability (Burns & Haller, 2015) and sexual orientation (Chung, 2007;Tagudina, 2012); research on candidate representation during elections (Fuchs & Schäfer, 2021;Heldman et al, 2018;Oates et al, 2019); and research on abuse on social media (Gorrell et al, 2018;MacAvaney et al, 2019;Waseem et al, 2017) and misand disinformation about public figures and other topics on social media (Guerin & Maharasingam-Shah, 2020;Sessa, 2020;Stabile et al, 2019). We used a broad definition of abuse for this study to cover 15 different types of online abuse based on social media reporting categories and prior research (Guerin & Maharasingam-Shah, 2020;Waseem et al, 2017).…”

Section: Manual Content Analysismentioning

confidence: 99%

“…This included a focus on a candidate's perceived or actual race, gender, sexual orientation, ethnicity, and religion. We relied on narrative categories from prior work (e.g., Oates et al, 2019) to understand if posters were more preoccupied with identity characteristics for some candidates over others. We found that tweets targeted at or about women of color were more likely to focus on their identity (6.2%) than white women (2.2%) or white men (0.4%).…”

Section: Tweets Targeted At Women Of Color Candidates Are More Likely...mentioning

confidence: 99%

See 1 more Smart Citation

An Unrepresentative Democracy: How Disinformation and Online Abuse Hinder Women of Color Political Candidates in the United States

Thakur¹,

Hankerson²,

Luria³

et al. 2022

Preprint

View full text Add to dashboard Cite

As more women have sought political representation by running for elected office, we have seen demonstrated increases in online harassment and abuse, including targeted mis- and disinformation campaigns. Researchers argue that these attacks are attempts to limit women’s ability to participate in electoral politics and suppress their voices in a variety of settings. While women in general may be subject to significant mis- and disinformation and abuse online, an intersectional approach which recognizes that women of color have to contend with multiple sources of oppression at the same time and that this impact is unique, can better illuminate the additional challenges faced by women of color in general and women of color political candidates specifically. We identified two key research questions: (1) Are women of color political candidates more likely to be subject to mis- and disinformation and online abuse compared to other types of candidates? (2) What are the impacts of mis- and disinformation and online abuse of women of color political candidates? We focus on both mis- and disinformation and online abuse because they are part of the larger problem of violence against women in politics and both are often aimed at undermining the political efficacy of women in public spaces.

show abstract

Section: Manual Content Analysismentioning

confidence: 99%

Section: Tweets Targeted At Women Of Color Candidates Are More Likely...mentioning

confidence: 99%

An Unrepresentative Democracy: How Disinformation and Online Abuse Hinder Women of Color Political Candidates in the United States

Thakur¹,

Hankerson²,

Luria³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Recent research shows the impact of online abuse on the formal involvement of women in politics. Oates et al (2019) analyse twitter responses to Democratic Party candidates for the 2020 U.S. Presidential primary election. They find that women candidates are frequently marginalized and attacked on character and identity issues that are not raised for their male counterparts.…”

Section: At the Societal Level Ogbv Impacts Media Freedommentioning

confidence: 99%

Global Evidence on the Prevalence and Impact of Online Gender-based Violence (OGBV)

Hicks¹

2021

View full text Add to dashboard Cite

This rapid review updates a previous report (Fraser and Martineau-Searle, 2018) with evidence from 2018 onwards. It finds an evidence base on online gender-based violence (OGBV) covering a wider range of countries than the previous report. Some key findings on the nature and prevalence of OGBV include: The most recent surveys show a prevalence of OGBV ranging from 16% to 58%; Men and boys also experience online abuse in high numbers, but it is less likely to be gender-based; Several studies from different countries identify Facebook as the top location for incidents of OGBV; Higher levels of online harassment and abuse are faced by people with intersecting inequality factors; According to victim-survivors, perpetrators are more likely to be unknown and acting alone, but large numbers are known to the victims. Perpetrators themselves report divergent, multifaceted and often over-lapping motivations for their actions; Analysis of underlying drivers of OGBV highlights an overarching theme of power and control, and heteronormative expectations around gender roles and sexual practice. Many authors recommend that OGBV be understood as part of a continuum of abuse where normalised behaviours, such as sexual harassment in public spaces, shade into behaviours widely recognized as criminal, such as physical assault. The societal impact of OGBV includes: Media freedom is compromised; Democracy being undermined; Economic losses resulting from lost productivity; A ‘climate of unsafety’ prevails. Evidence base: The number of surveys about self-reported experiences with online harassment has increased rapidly. The majority of the research found during the course of this rapid review came from international and domestic non-governmental organisations and think-tanks. Academic research studies were also found, including several literature reviews.

show abstract

“…Oates, S., Gurevich, O., Walker, C., & Di Meco, L. (2019). Running While Female: Using AI to Track how Twitter Commentary Disadvantages Women in the 2020 U.S. Primaries (SSRN Scholarly Paper ID 3444200).…”

mentioning

confidence: 99%

Facts and their Discontents: A Research Agenda for Online Disinformation, Race, and Gender

Thakur¹,

Madrigal²

2022

Preprint

View full text Add to dashboard Cite

The January 6th attack on the U.S. Capitol demonstrated how online disinformation can have severe offline consequences. For some time, the problems and possible impacts on democracy caused by online mis- and dis-information have dominated public policy discussions and thus research about these topics has developed rapidly in the last few years. However, this research generally lacks a focus on the impact of disinformation and misinformation on people of color, women, LGBTQIA+ communities, and other voices that are less prominent in mainstream political discourse in the U.S.Many disinformation campaigns are specifically designed with racist and/or misogynistic content, suggesting that disinformation is a tool used to promote ideologies like white supremacy and patriarchy.In September 2020, CDT brought together an interdisciplinary and international group of experts to share and discuss research on this issue. This report presents some of those ideas and builds upon them to identify key research opportunities, including important unresolved questions around the intersections of online disinformation, race, and gender. This report also makes recommendations for how to tackle the related methodological and technical problems that researchers and others face in addressing these topics. This is important in generating research that will be directly relevant for developing policy solutions to address disinformation.

show abstract

Running While Female: Using AI to Track how Twitter Commentary Disadvantages Women in the 2020 U.S. Primaries

Cited by 8 publications

References 21 publications

An Unrepresentative Democracy: How Disinformation and Online Abuse Hinder Women of Color Political Candidates in the United States

An Unrepresentative Democracy: How Disinformation and Online Abuse Hinder Women of Color Political Candidates in the United States

Global Evidence on the Prevalence and Impact of Online Gender-based Violence (OGBV)

Facts and their Discontents: A Research Agenda for Online Disinformation, Race, and Gender

Contact Info

Product

Resources

About