[Colloq] Hiring Talk: Estimating Ideology Using Facebook’s ‘Like’ Data - Robert Bond - February 19th 12:30pm 366 WVH

Jessica Biron bironje at ccs.neu.edu
Wed Feb 13 15:28:49 EST 2013


Estimating Ideology Using Facebook’s ‘Like’ Data 
Robert Bond (UCSD) 




Tuesday, February 19th 

12:30 – 2:00 pm 

West Village H – Room 366 




(A light lunch will be served.) 




Abstract: 


I develop a statistical model to estimate the ideology of politicians and their supporters using Facebook data about which users publicly support which political figures by ‘liking’ them on the site. This process produces ideological estimates for both political pages on Facebook and the users that like those pages. The ideological estimates that result from the scaling process for the political pages correlate highly with DW-NOMINATE. The results for Facebook users are predictive of their self-reported political views. I then use these measures of ideology to study the structure of ideology in social space. Indeed, I find that ideology is highly correlated among friends and family and that closer friends are more similar ideologically. I study how the social structure of ideology has changed over time. The results show that in 2012 close friends are more similar ideologically than in 2011, suggesting an increase in polarization. Finally, I examine how the ideology of an individual's social contact affects the propensity to vote. 




Bio: 

Robert Bond received his B.A. in Political Science from Arizona State University and is a Ph.D. candidate in Political Science from the University of California at San Diego. He uses computational approaches to study political phenomena using large-scale datasets. His dissertation examines the structure of ideology and patterns of voter turnout in our social networks. Much of his work examines large-scale datasets derived from online social networks (such as Facebook and Twitter) to refine an understanding of the effects of our social network on how we think, feel and behave politically. His work has been published in the Nature, PLoS One, and American Politics Research. 



Jessica Biron 
Administrative Assistant – Office of the Dean and CCIS Development 
College of Computer and Information Science 
Northeastern University 
202 West Village H 
617-373-5204 
bironje at ccs.neu.edu 
http://www.ccs.neu.edu/ 


More information about the Colloq mailing list