Ongoing work at the Social Science Research Council (SSRC) over the past several years identified the importance of access to socially relevant data in order for researchers to address large issues that enhance scholarly knowledge and also inform, empower, and protect the public interest. However, as noted in the SSRC’s 2018 “To Secure Knowledge” report, while in the past much of data used by social researchers was gathered by governments and nonprofits, today much of the most relevant data is held by private companies. Although social media platforms are beginning to engage the broader research community, critical questions of public importance—such as the role of social media in democratic politics—need to be addressed now with both new and existing approaches that meet the highest scientific and ethical standards.
The Social Data Research and Dissertation Fellowship program, offered by the SSRC Social Data Initiative and supported by the Omidyar Network, seeks to encourage multifaceted pathways for the collection and analysis of social data, with the larger aim of cultivating robust research on technology and society. Multiple pathways for journalistic and academic study of proprietary socially relevant data are necessary in order to build a robust culture of research and accountability and therefore better protect democracy and the public.
In this initial grant cycle, fellows were asked to focus on the forthcoming 2020 election in the United States, while thinking creatively about possible approaches to gathering and analyzing relevant data. In selecting the fifteen projects below, the SSRC followed the values that have guided it in its nearly 100-year history, including a commitment to an open application process that invites all eligible proposals and adherence to a rigorous, independent peer review process that evaluates projects based on clearly stated review criteria. Because of ongoing concerns about privacy and ethics related to social data, the SSRC developed and applied an additional ethics review process for all recommended proposals. This ethics review process is intended to serve as an additional check and, most importantly, provide substantive feedback to successful applicants so that they can refine their projects with an ethical framework in mind.
The selected projects will offer important and timely perspectives to the 2020 election. From studying the impact of inequality on political participation to exploring the prevalence and effects of mis- and disinformation, the awarded projects also reflect a full range of methodological and disciplinary approaches: experiments, computational and panel surveys, content analysis, and ethnographic interviews, as well as computational text analysis, sentiment analysis, and other natural language processing approaches to social media datasets. Projects will strengthen scholarly understanding of how polarization, toxicity, and inequality impact the election. A grant to Social Data Research Fellow Brooke Foucault Welles supports a project that, drawing on Twitter and other unique research data, will examine how marginalized groups experience political voice, attention, and participation in the United States, and how that may translate to political behavior in the 2020 election. Social Data Research Fellow Ashley Muddiman will explore—through quantitative content analysis and an experiment—patterns of incivility targeting women candidates during the 2020 general election. Other awarded projects examine the ethics of social media research, trust and skepticism in the media, and the effect of anonymity on political—and toxic—speech, among many other topics.
Each project also brings unique comparative, cultural, and historical context to the study of social data drawn from a wide range of platforms, including common social media platforms like Twitter, reddit, and Facebook, streaming video platforms such as YouTube and Twitch, recommendation systems from Amazon, and even gaming platforms like World of Warcraft. Social Data Research Fellow Jennifer Evans will analyze how historical analogy is deployed to normalize misinformation in both the Covid-19 crisis and the 2020 US election, while Crystal Lee, a Social Data Dissertation Fellow, explores how visualizations and graphics circulated through social media can both serve the public good and be manipulated to spread misinformation. While many projects focus on national politics, several also emphasize local and state elections, including Dissertation Fellow Stephanie Ternullo’s comparative study of three midwestern towns.
These examples represent just a few of the projects underway with support from the Omidyar Network. Overall, the fifteen projects represent nearly $500,000 in funding that will further our understanding of how social data and media platforms play a daily role in our lives, expand the scholarly methodological toolkit, and generate valuable datasets and findings for future research.
The following are awardees from the Social Data Research and Dissertation Fellowship program: