Will Hobbs

I develop and apply statistical methods to explain political and social phenomena. My work focuses on the social effects of government policies, on how small groups of people adapt to sudden changes in their lives, and on low-dimensional representation of social interaction and language. Most of the applications are in American politics, public policy, and health care.

For example, some of my current research studies public attitudes on the Affordable Care Act. To measure attitudes before and after implementation of the ACA and to get around limitations of multiple choice survey questions, I have developed a method to summarize latent meaning in open-ended survey text about the law. The method combines survey text with general English word meanings to estimate the gist of survey responses in just a few dimensions of meaning. It outperforms existing approaches because it performs well on vague and keyword based text.

My past work has been published in the American Journal of Political Science, the Journal of Politics, and the Proceedings of the National Academy of Sciences, among other outlets. It has also been covered in popular media, including The Atlantic, Science Magazine, The New York Times, the Los Angeles Times, and The Wall Street Journal.

From 2016 to 2018, I am a postdoctoral fellow at Northeastern University's Network Science Institute. I earned my PhD in political science from the University of California, San Diego.

Working Papers

Pivoted Text Scaling for Open-Ended Survey Responses
William Hobbs.

How Sudden Censorship Can Increase Access to Information
William Hobbs and Margaret Roberts.
(conditionally accepted, American Political Science Review)

Effects of the 2016 Presidential Campaign on the Day-to-Day Segregation of Arab and Muslim Americans
William Hobbs and Nazita Lajevardi. (email for working paper)

Partisan Attachment or Life Stability?
William Hobbs.
(conditionally accepted, Political Behavior)

Publications (with links to papers)

ConStance: Modeling Annotation Contexts to Improve Stance Classification
Kenneth Joseph, Lisa Friedland, Oren Tsur, William Hobbs, and David Lazer. 2017.
Empirical Methods in Natural Language Processing (EMNLP) '17.

Voters of the Year: 19 Voters Who Were Unintentional Election Poll Sensors on Twitter
William Hobbs, Lisa Friedland, Kenneth Joseph, Oren Tsur, Stefan Wojcik, and David Lazer. 2017.
AAAI Conference on Web and Social Media (ICWSM) '17. 544-547.

Connective Recovery in Social Networks After the Death of a Friend
William Hobbs and Moira Burke. 2017.
Nature: Human Behaviour. 1 (5).
-- Science
-- The Atlantic
-- UC San Diego press release

Online Social Integration is Associated with Reduced Mortality Risk
William Hobbs, Moira Burke, Nicholas Christakis, and James Fowler. 2016.
Proceedings of the National Academy of Sciences. 113 (46): 12980-12984.
-- New York Times
-- Los Angeles Times
-- UC San Diego press release
-- Altmetric 2016 Top 100

Presidential Effort and International Outcomes: evidence for an executive bottleneck
David Lindsey and William Hobbs. 2015.
Journal of Politics. 77(4): 1089-1102.
-- supporting information
-- replication materials

Widowhood Effects in Voter Participation
William Hobbs, Nicholas Christakis, and James Fowler. 2014.
American Journal of Political Science. 58(1): 1-16.
-- supporting information
-- replication code (email me for updated code in Python)
-- The Atlantic

Prediction of Mortality Using On-line, Self-Reported Health Data: empirical test of the RealAge score
William Hobbs and James Fowler. 2014.
PLOS ONE. 9(1): e86385.
-- supporting information

Embedded Interests and the Managerial Local State: the political economy of methanol fuel-switching in China
Genia Kostka and William Hobbs. 2013.
Journal of Contemporary China. 22(80): 204-218.

Local Energy Efficiency Policy Implementation in China: bridging the gap between national priorities and local interests
Genia Kostka and William Hobbs. 2012.
China Quarterly. 211: 765-785.