2013-14: CMSS: Social Network Data: Collection and Analysis: Oct. 21-23, 2013


A two-and-a-half day workshop, held at SAMSI, that involved roughly 12-15 invited talks along with formal panel discussions, a poster session, and time available for informal collaboration-building discussions.


This workshop directly interfaced with the Computational Methods in Social Science program year by focusing on pressing issues in the systematic collection, statistical analysis, and mathematical modeling of social science network data. The social world is inherently one of interacting entities. While qualitative and theoretical social science has long had free reign to study complex structures arising from the relations among multiple entities, recent advances in network statistics have begun to allow for the quantitative exploration of these more complex network structures that are central to the structure of the social world. Fundamentally, all networks consist of nodes and edges, or relations between those nodes. Perspectives on networks and the possibilities for statistical research based on such structures are myriad and varied. Additional dimensions of data may be available, including: flows over edges, dynamics over time, and static or fixed covariates on any of the above. Inferential perspectives can then aim to characterize any sub-set of these variables either jointly or conditioning on any others. Data collection, sampling, experimentation, and missing data add further levels of complexity. While the methodological questions associated with networks are broad and disparate, so are the substantive problems they are able to address. Indeed, it is these substantive problems that determine which statistical problems are addressed first. By focusing on data collection efforts (e.g. Add Health, micro-financing in Indian villages) and the relevant methodologies for their analysis, we aimed to further engage mathematical, statistical and computational approaches with social science questions.

Schedule and Supporting Media

Participant List
Speakers, Titles and Abstracts
Poster Titles

Monday, October 21, 2013

Time Description Speaker Slides Videos
9:00-9:30 Registration
9:30-9:40 Welcome Remarks (Peter Mucha, University of North Carolina)
9:40-10:15 Estimating Network Degree Distributions from Sampled Networks: An Inverse Problem Eric Kolaczyk, Boston University
10:15-10:50 Inference from Link-Tracing Network Samples Krista Gile, University of Massachusetts
10:50-11:20 Break
11:20-11:55 Mixed Membership of Experts Stochastic Blockmodel Brendan Murphy, University College Dublin
11:55-1:55 Lunch (at SAMSI)
1:55-2:30 Cultural Enrichment: Linking Structure to Culture in Network Analysis Jacob Foster, UCLA
2:30-3:05 Latent Space Models for Multiview Network Data Tyler McCormick, University of Washington
3:05-3:35 Break
3:35-4:10 Asking Questions about Numbers: Practical Considerations in RDS Degree Measurement Elena Erosheva, University of Washington
4:10-5:00 Student Poster Fast Forward
5:00-7:00 Poster Session and Reception

Tuesday, October 22, 2013

Time Description Speaker Slides Videos
9:00-9:30 Registration
9:30-10:05 Tracking Influence in Dynamic Social Networks Rebecca Willett, University of Wisconsin
10:05-10:40 Local Clustering and the Blessing of Transitivity Karl Rohe, University of Wisconsin
10:40-11:10 Break
11:10-11:45 Differentially Private Graphical Degree Sequences and Synthetic Graphs Aleksandra Slavkovic, Pennsylvania State University
11:45-1:45 Lunch (at SAMSI)
1:45-2:20 Protocols for Randomized Experiments to Identify Network Contagion A.C. Thomas, Carnegie Mellon University
2:20-2:55 Graph Cluster Randomization: Design and Analysis for Experiments in Networks Johan Ugander, Cornell University
2:55-3:25 Break
3:25-4:25 Panel Discussion: Tom Carsey, UNC;
Steve Fienberg, CMU;
Mark Handcock, UCLA
6:00 Dinner on your own in small groups at the Streets at Southpoint mall

Wednesday, October 23, 2013

Time Description Speaker Slides Videos
9:00-9:30 Registration
9:30-10:05 Topic-Partitioned Multinetwork Embeddings Bruce Desmarais, University of Massachusetts
10:05-10:40 Is Intermediate-Scale Structure Tree-like in Social Networks? Blair Sullivan, N.C. State/ORNL
10:40-11:10 Break
11:10-11:45 When Can We Learn Network Models from Samples? Cosma Shalizi, Carnegie Mellon University
11:45-11:55 Concluding Remarks
11:55-1:55 Box Lunch (at SAMSI)