Course objectives:

The objective of the course is to provide the PhD student with both theoretical and technical knowledge and competences to understand what social network analysis is, what type of network analysis methods are available, and how this method can be used in their context of research and integrate in their PhD theses. The course should provide a comprehensive knowledge of fundamental concepts that stem from Social Network theory and its implementation in specific research context through the use of network visualization and network analysis (network modelling) tools.

The PhD students will have the opportunity to learn how to use different software, such as Visone, VNet and MPNet for both network analysis and visualization.

The course will gradually increase in its complexity where it will begin with broader discussions about general idea and concepts of networks, brokerage, structural holes etc., and through the discussions about network characteristics (cohesion, connectivity etc.), nodes and attributes, arrive to a more complex discussion on the use of exponential random graph models (ERGMs), which will ultimately require from the students to engage their own data (real or hypothetical), in order to have hands on the procedure required to prepare and analyze the data prior to producing the presentation of their findings.

The course dynamics:

The dynamics of the coursework will be based on short and efficient sequences of about 30min lectures and 30 min follow-up exercises on previously presented topic. In order to give the most efficient learning process, the course will span across two weeks, two days each week, in order to enable the participants to have enough time to get familiar with the course material and the software.

Course program (detailed):

Day 1: Introduction to Social Network Analysis:

• Fundamental network concepts – networks: what and why?

• Thinking about networks: Research question and research design

• Social systems and network structures: Relational ties and actor attributes

• Graphs, paths, adjacency matrices

• Different methods for conducting SNA (qualitative and quantitative)

• Data importing, analysis, and visualization

The aim of the first day is to introduce the students to the main conceptual and theoretical framework of social network analysis in broader sense. The students will learn and reflect upon the idea of SNA and its potential applicability in various contexts of research within the field of social science. The purpose of the first day is to set the stage for the coming days, in which the students will begin to reposition the idea of their own research design into a possibility to understanding their problems and research questions from the angle of social networks.

Throughout the day the main concepts which will be in use during the rest of the course will be introduced, and we’ll also intend to provide examples where SNA has been used in different fields of social science research to motivate and inspire rethinking of the participants’ existing research designs. Also, the students will have the opportunity to delve into methodological aspects of SNA, including data collection and data analysis (how and why). Finally, they will explore network visualization software and conduct basic descriptive analysis of various preset network datasets.

Day 2: Network Visualization and introduction to ERGMs:

• One-mode and two-mode networks

• Density, reciprocity, and clustering

• Degrees and centrality

• Cohesion, connectivity, and community

• Network configurations and actors’ attributes

• Association between structures and attributes

From the second day, participants will have the opportunity to work with actual network data and different software solutions. This hand-on experience will enable them to conduct network analysis using real network data. During this day, students will have the learn about one- and two-mode networks, as well the internal processes that lead to establishing different modules for interpretating network structures. The first two days will primarily focus on descriptive network analysis and visualizations, as well to getting familiar for working with network data.

Day 3: ERGMs for one-mode network

• ERGMs

• Network simulation and network estimation

• Estimating social circuit models

• Goodness of fit

The third and the fourth day will be primarily dedicated to network modelling, focusing on measuring network parameters for statistical inferencing working with one-mode network data. Participants will have the opportunity to learn how actual network data is modelled using MPNet software. During this day, the primary focus will be on modeling network parameters in one-mode networks and testing the goodness of fit of the resulting network statistics as a part of robustness check.

Day 4: ERGMs for two-mode networks

• Network variables: endogenous and exogenous

• Actor attributes

• Bipartite networks

• Multilevel ERGMs

On the last day of the course, we aim to provide an overview of the most complex network statistical modeling, including bipartite and multilevel networks. This is important as some research design would require using two different groups of social actors and inclusion of different types of ties. Also, during this day, students will learn the importance of actors’ attributes on network tie formation process. Lastly, they will gain an understanding of how actors’ attributes influence the formation of other actors’ attributes within the network.

Disclaimer:
DDSA has explicit permission from Arcanic and the owners of the https://phdcourses.dk/ website to display the courses on ddsa.dk.