đź“„ View the full conference program:
Click here to open the program
đź“ť Help us plan!
Please complete a short survey to indicate your interest in space-limited events like mentoring sessions and informal meet-ups:
Click here to fill out the survey
We’ll kick things off with two days of hands-on training workshops at Northeastern University, taking place August 11–12, just ahead of the main conference.
Introduction to Network Analysis in R
Michael Heaney
This workshop introduces participants to the fundamentals of conducting network analysis using R. It begins with a discussion of data formats and procedures for importing data, including whole network and ego network data. It moves to accounting for subgroups and configurations present in the data, such as cliques and triads. Substantial attention will be given to centrality and brokerage. If time allows, additional topics such as community detection may be discussed. No prior experience with R is assumed. However, a basic conceptual understanding of networks may be helpful (e.g., what are links, nodes, modes?).
Network Visualization with R
Katya Ognyanova
This workshop covers network visualization using the R language for statistical computing (cran.r-project.org) and RStudio (rstudio.com). Participants should have some prior knowledge of R and network concepts. The workshop provides a step-by-step guide, using a series of examples, to demonstrate the path from raw data to graph visualization in the igraph and Statnet frameworks. The advanced portion of the workshop will outline the use of dynamic visualization for longitudinal networks and the integration of networks with geographic maps. Additionally, we will explore methods for converting R-generated graphs into interactive JavaScript-based visualizations for the web.
Introduction to Latent Variable Models for Networks
Shahryar Minhas
This workshop offers an introduction to latent variable models, with a focus on latent factor models (LFMs) for network analysis in R. You'll see how latent variable models can serve as a regression tool for dyadic data and as a measurement device that place actors in a "social space" based on their interaction patterns. We'll demonstrate how to implement these models in R and interpret the results. We'll also discuss extensions to longitudinal networks and key considerations for making valid inferences. Basic familiarity with social network concepts, regression, and R will be assumed.
Introduction to Exponential Random Graph Models (ERGMs)
Cassie McMillan
This workshop introduces participants to exponential random graph models (ERGMs), a multivariate statistical technique for social network analysis. ERGMs are commonly applied to study how actor-level characteristics, features of dyads, and structural patterns inform networks of interest to political scientists. The workshop will begin with an introduction to the statistical methodology of ERGMs that focuses on the model’s applicability for addressing social science research questions. Then, it will provide an overview of how to implement the models in R, interpret their results substantively, and assess their goodness of fit. By the end of the workshop, participants will have hands-on experience estimating ERGMs on empirical network datasets. A basic understanding of social network concepts and data management is assumed. Participants will also get the most out of this workshop if they have familiarity with logistic regression and previous experience with R.
Introduction to the National Internet Observatory
David Lazer and Pranav Goel
Today, researchers face significant ethical and technical barriers when attempting to collect data on online behavior and Internet activity for academic studies. Major social media platforms like Facebook and Twitter have progressively restricted access to their official Application Programming Interfaces (APIs), which previously served as primary tools for researchers to create customized datasets from specific platforms for studying online content production and engagement behavior. This tutorial introduces the National Internet Observatory (NIO), an alternative data collection framework and infrastructure designed to help researchers study online behavior, with a particular focus on content viewing— the predominant form of online activity. This data framework is created by academics to enable novel cross-disciplinary and cross-platform academic research across web and social media environments. The tutorial presents NIO's informed data donation process, types of data collected, information about the participant sample and the structure as well as examples of analyses and research with the trace data, and an overview of the application process and the secure computing infrastructure. We will also discuss certain datasets in-depth, and enable hands-on exploration of the data to demonstrate how NIO can enable your research in the post-API era. If you are interested in this workshop in particular, please click here.