NYU Politics Data Center

2014 Northeast Political Methodology Meeting at NYU

Registration closes April 28, 2014.

 

All events take place at the NYU Politics Department, Room 217.

19 W 4th Street, 2nd Floor (Get Google Maps directions here)

(Corner of W 4th Street and Mercer Street)

 

Friday, May 02, 2014

 

11:30 - 12:30:   Lunch

 

12:30 - 01:30:   Bruce Desmarais

Department of Political Science, U-Mass Amherst

Hanna Wallach

Microsoft Research NYC and Umass Amherst

"Communication Network Content and Structure: A Modeling Approach with Application to Gender Mixing in Local Government Internal E-Mail Communication"

                          [Abstract]We develop and implement a statistical model that jointly (1) partitions text-valued communication networks into sub-networks by the topic of communication and (2) infers a probabilistic model of the sub-networks' structures. The probabilistic graph models represent common covariate effects in addition to capturing residual relational dependencies. Specifically, the methodology we develop integrates latent Dirichlet allocation for the analysis of text with latent space graph projection for representing network structure. This model is applicable to many real-world social, organizational and political communication networksds. We present an application to local government internal communication data. We model email corpora from four North Carolina county governments. In modeling network structure, we draw upon the organizational communication literature on gender bias. We study how patterns of gender mixing vary with the topic of communication within county government e-mail networks. [Paper]

 

01:45 - 02:45:   Elizabeth Ogburn

Department of Biostatistics, Johns Hopkins University

"Causal and Statistical Inference for Network Dependent Data"

                          [Abstract]Increasing interest in and availability of network data necessitates new methods for causal and statistical inference when observations are linked by network ties. My talk is motivated by the Health Outcomes, Progressive Entrepreneurship, and Networks (HopeNet) Study, which will collect three waves of complete social network data and implement clean water and microenterprise interventions in a small community in southwestern Uganda. Causal effects of interest include the effects of an individual's exposure to each intervention on his own outcome, and several different types of effects of an individual's exposure on the outcomes of his social contacts. In order to clearly articulate these latter “interference” effects, I differentiate three different causal mechanisms that give rise to interference, defined as an effect of one individual's exposure on another's outcome, and briefly discuss new identification results for interference effects. I then turn to the problem of estimation when only a single network of non-independent observations is observed and the dependence among observations is informed by network topology. I explain why results on spatial-temporal dependence are not immediately applicable to this new setting and present some new methods for estimation in the presence of network dependence. [Paper]

 

03:00 - 04:00:   Erin Hartman

Bluelabs

Design Based Approaches for Unit Non-Response

 

04:15 - 05:15:   Brian Keegan

Northeastern University

"Get Back! You Don't Know Me Like That: The Social Mediation of Fact Checking Interventions in Twitter Conversations"

                          [Abstract]The prevalence of misinformation within social media and online communities can undermine public security and distract attention from important issues. Fact-checking interventions, in which users cite fact-checking websites such as Snopes.com and FactCheck.org, are a strategy users can employ to refute false claims made by their peers. While laboratory research suggests such interventions are not effective in persuading people to abandon false ideas, little work considers how such interventions are actually deployed in real-world conversations. Using approximately 1,600 interventions observed on Twitter between 2012 and 2013, we examine the contexts and consequences of fact-checking interventions.We focus in particular on the social relationship between the individual who issues the fact-check and the individual whose facts are challenged. Our results indicate that though fact-checking interventions are most commonly issued by strangers, they are more likely to draw user attention and responses when they come from friends. Finally, we discuss implications for designing more effective interventions against misinformation. [Paper]

 

05:30 - 06:30:   Post Papers Discussion

 

06:45                Dinner for Speakers and Invited Faculty Guests