Network Analyses

COURSE LEADER 

Edwin de Jonge

TARGET GROUP 

Staff of NSIs interested in advanced methodological tools. 

Analysts working on skills mapping, organizational collaboration, data integration, or innovation projects. 

Statisticians and data scientists analysing complex systems, relationships, and interactions (e.g., trade, migration, social networks) 

ENTRY 

QUALIFICATIONS 

• MSc level or higher (or equivalent) 

• Basic statistical knowledge 

• Experience with Python 

OBJECTIVE(S) 

• Understand the fundamental principles of network analysis and graph theory. 

• Apply quantitative network metrics (centrality, clustering, modularity, etc.) to measure structures and relationships. 

• Learn which modern software tools (R/Python) are used for building and visualizing networks. 

• Learn about state-of-the-art Official Statistics examples 

• Use network models in real-world contexts 

• Evaluate how network analysis can support modernization of official statistics by learning from current examples. 

CONTENTS 

The Network Analyses three-day course provides an introduction to network science tailored to the needs of official statistics. Participants will learn the fundamental concepts of networks, including nodes, edges, and key properties such as centrality, clustering, and modularity. Through practical sessions Python, they will gain handson experience in building, analyzing, and visualizing networks. The programme covers advanced topics such as community detection, statistical models for networks, and the challenges of storage and computation. Special attention is given to applying network methods to real-world statistical domains, including transportation, trade, migration, and enterprise data. Case studies from national statistical institutes illustrate both the opportunities and limitations of network approaches in practice. Ethical and confidentiality considerations are also discussed to ensure responsible use of network data. The course concludes with group projects where participants construct, analyze, and present network pipelines based on real datasets.   

EXPECTED 

OUTCOME 

• Understanding fundamentals of applied network science 

• Awareness of opportunities of network science for Official Statistics 

• Know how to write Python code to implement networks and networks analysis. 

        -     Improved knowledge on network storage, 

TRAINING METHODS 

Lectures with slides, hands-on exercises, discussions and interactive assignments. 

REQUIRED READING  

n/a 

SUGGESTED READING 

• Newman, M.E.J. “Networks: An Introduction”.  

• Barabasi, A. “Network Science” 

• Van der Laan, J., De Jonge, E., Das, M., Te Riele, S., & Emery, T. (2023). A whole population network and its application for the social sciences. European sociological review, 39(1), 145160. 

• de Jonge, E., Pijpers, F. P., & Mandjes, M. (2025).  Deriving production chains using restricted gradient extraction. Chaos:  

An Interdisciplinary Journal of Nonlinear Science, 35(5). 

REQUIRED 

PREPARATION 

Python installation + github repo (to be provided) 

TRAINER(S)/ 

LECTURER(S) 

Edwin de Jonge (CBS Netherlands)

Jan van der Laan (CBS Netherlands)