Computational Social Science Forum
Date: Monday, April 12, 2021
Time: 12:00-1:30 PM Pacific Time
Location: Register to receive the schedule and access links.
Decoding the Digital Gender Gap: Online Gender Bias is Stronger in Images than Text
Women are systematically underrepresented in high status occupations and social categories, constituting a pervasive gender gap in society. What is more, recent advances in computational linguistics show that biases in gender representation are deeply woven into the texts that people consume, produce, and exchange on a daily basis, such as books, print media, and online blogs. However, text is only one of the many modalities through which cultural knowledge is exchanged. Human communication is also embodied and involves numerous image-based approaches to encoding and exchanging cultural associations. We predict that online images will encode a greater bias in gender representation than text, since images often depict the embodied person, where visual cues readily elicit the identification of gender. By contrast, in text, it is possible to describe people generically without specifying gender, e.g., by referring to someone as the doctor or the banker. By comparing leading techniques for automatically detecting gender bias in text and images, our results suggest that text-based measures of gender stereotypes may significantly underestimate the gender gap in society.
The Computational Social Science Forum is an informal setting for the interdisciplinary exchange of ideas and scholarship at the intersection of social science and data science. Weekly meetings are hosted by researchers from BIDS and D-Lab, and participants engage in a variety of activities such as presentations of work in progress, discussions and critiques of recent papers, introductions to new tools and methods, discussions around ethics, fairness, inequality, and responsible conduct of research, as well as professional development. We welcome social scientists researchers with interests in data science methods and tools, and data scientists with applications or interests in public policy, social, behavioral, and health sciences. Participants include graduate students, postdocs, staff, and faculty, and members are encouraged to attend regularly in order to foster community around improving computational social science research, supporting the development and research of group members, and fostering new collaborations. This Forum is organized as part of the Computational Social Science Training Program, and interested UC Berkeley community members are invited to use this registration form to receive the schedule and access links. Please contact email@example.com for more information.
Douglas Guilbeault is an Assistant Professor in the Management of Organizations Group at the Haas School of Business. He studies how people learn, challenge, develop, and invent categories by communicating in social networks. This investigation extends to the analysis of how organizations mediate and augment social computation by enabling new forms of communication, coordination, and creativity. This investigation further extends into how the social construction of meaning can be shaped by various sources of influence, such as political messaging and the design of social media platforms. His work on these topics has appeared in a number of journals, including Nature Communications, The Proceedings of the National Academy of the Sciences, Cognition, Policy and Internet, and The Journal of International Affairs, as well as in popular news outlets, such as The Atlantic and Wired. Guilbeault’s work has received top research awards from The International Conference on Computational Social Science, The Cognitive Science Society, and The International Communication Association. In addition, he was a recipient of Stanford’s “The Art of Science” award for the piece “Changing Views in Data Science over 50 Years” produced in collaboration with the research collective, comp-syn. Guilbeault teaches People Analytics at Haas, focusing on how organizations are using novel algorithmic methods to address (and sometimes inadvertently create) fundamental problems in management.