I'm an Assistant Professor of Sociology at Rutgers University in New Brunswick, New Jersey. I recently completed my PhD in Sociology at Cornell University, where I was a member of the Social Dynamics Lab. My research and teaching interests are in political sociology and social movements, culture, and social networks. I specialize in using computational methods including social network analysis, machine-learning, and natural language processing. My work has been published in venues including Social Forces, Mobilization, and Socius.
Social media technologies have transformed both politics and our ability to study it. In my dissertation research I pose three related questions, drawing upon theories from political sociology, social movements, and public opinion. Why have far-right, populist actors have been so successful at building remarkably large online audiences and has this online presence contributed to their electoral performance? To what extent is online social movement activity, engagement, and recruitment associated with offline events like protests, elections, and terrorist attacks? What is the role of individual opinion leaders in massive online debates? To address these questions, I have compiled novel datasets combining information from social media, newspapers, and other sources, which enable me to study the dynamics of activism and debate at scale and with high granularity. Methodologically, I use a range of statistical and computational methods including social network analysis, natural language processing, and time series modeling. These studies contribute to our understanding of contemporary far-right politics, online debates, and public opinion on issues including immigration and Brexit.
A second area of my research focuses on identifying and understanding hate speech and hate speakers on social media. You can read our ICWSM paper on automated hate speech detection here. Our work has been covered in Wired Magazine, Tech Republic, and New Scientist. We also wrote a paper focusing on the commonalities between hate speech detection and other topics, including online abuse and cyberbullying, for the Association for Computational Linguistics 1st Workshop of Abusive Language Online. In 2019 my coauthors and I conducted a study on racial bias in hate speech and abusive language detection datasets, published in the proceedings of the 3rd Workshop for Abusive Language Online. We demonstrate that machine-learning classifiers trained on several widely used datasets are more likely to predict that tweets written in African-American English are hateful than similar tweets written in Standard American English. You can read the paper here, along with coverage in Vox.
I collaborated with Paromita Sanyal to study the impact of microcredit participation on women's social networks in rural India. We found that participation in associations allows women to expand their personal networks and enhance their social capital. Our work was published in Social Forces. I am currently working on a project with my colleague Antonio Sirianni to study the career trajectories of adult film performers and another solo-authored project to study how co-performance and co-production shape the development of musical genres.
In addition to my substantive interests, I also study how computational methods can be applied more generally in sociological research. For example, as part of the Fragile Families Challenge, I examined how neural networks can enable us to predict social outcomes and assessed the extent to which these black box predictive models can be amenable to sociological explanations. I found that these methods do not radically outperform traditional approaches like linear regression, but may allow us to use large amounts of data to inductively identify important variables. A paper based on my analysis is forthcoming in Socius (pre-print) and our paper describing the results of the Fragile Families Challenge mass collaboration was recently published in PNAS.
In addition to my academic research I have extensive experience using computational methods in industry settings. In the summer of 2016 I was an Eric and Wendy Schmidt Data Science for Social Good Fellow at the University of Chicago. I worked on a project to develop an early-warning system to identify police misconduct and helped develop a new model to predict risks at the dispatch level. You can read about our work here, along with media coverage in The Chicago Tribune, NPR, Mother Jones, the Economist, and Forbes. As of spring 2017 our system is being implemented into two police departments, you can read about the progress here. In 2017 I spent the summer in the Data Science Research & Development group at Civis Analytics, a data science consulting and software company based in Chicago. I used natural language processing and machine learning techniques to build a tool to monitor political discussion on Twitter. In 2018 I was a Core Data Science intern at Facebook in Menlo Park. I conducted a study of misinformation sharing on the platform and helped to evaluate and deploy a new tool related to their on-going election integrity efforts.
Please feel free to get in touch at thomas dot davidson at rutgers dot edu.