​Social Network Analysis with R 


27 - 29 January


Target audience 

The course caters for two types of participants. You are either in a data-analytical field and you want expand your repertoire to social network analysis. Or you have a theory-heavy background and would like to learn how to use R to analyse social network data. The course assumes prior experience with R, or python (with a fast bridging course in R prior to starting). Whether you are particularly interested in Social Network Analysis, or would simply like to be able to apply the “network lens” to unlock more insights, this course is appropriate. The course is accessible, but participants can easily scale to more advanced outcomes.


First, we introduce the strong theoretical intuitions that have guided the field since the 1930s. This is done through tracing the origins in structuralism, telling the story of the sociologists' dark ages and the recent entrance of the physicists. Key theories such as the strength of weak ties, structural holes, the small world problem, and social capital are discussed. The result should be an intuition for identifying network structures in data, and how network based measures can be interpreted in different contexts.

Second, introduce the building blocks to analyse network data, primarily appropriate data structures and how to coerce everyday datasets to uncover network structure. This includes coercing data into and from square adjacency matrices, incidence matrices, edgelists with the help of the popular igraph library in R.

Third, we cover the highlights of network analytics such as network position, network structure, and community detection.

​The whole workshop will utilise Twitter datasets, since these offer non-trivial network structures, as well as text data and relatively interpretable grounding for measurements.​

Expected outcomes 

  • Create, import and export network data
  • Analyse network level, node level and substructure metrics of network data
  • Learn to apply community detection and various wrangling procedures
  • Learn to visualise network data in meaningful ways
  • Recognise network data in non-obvious places and extract unique insights
  • Learn to collect and analyse social media data, particularly Twitter data
  • Learn how to run the entire workflow on a remote compute environment in the cloud with AWS.

Course format 

There will be lectures, interwoven with practical exercises. 
Participants will be encouraged to follow along with exercises and programming on their own laptops.  


Participants are required to have R, Rstudio and Gephi installed prior to the course.  

The presenter 

Dr. Laurenz Cornelissen specialises in social network analysis, particularly expanding the application of network data to fields such as bot-detection and social media data. He holds a PhD in Decision-making and Knowledge Dynamics and started the Computational Social Science Research group at the Department of Information Science at Stellenbosch University. He is a social scientist who self-converted to something mimicking a computer scientist, he has a rich theoretical grounding in organisational and social science and is very enthusiastic about the application of computational methods.