CONNECTING DOTS: Social Network Analysis with R*

The Social Sciences Research Laboratories (SSRL) is pleased to offer two world-class interactive online workshops on social network analysis, scheduled for September 2 to 11, 2020. The first workshop focuses on Applying Social Network Analysis with R, while the second workshop focuses on Statistical Analysis for Social Network Data (see detailed descriptions below).

The workshops are led by Filip Agneessens, a leading international scholar in the field of social network analysis. Filip is an Associate Professor in the Department of Sociology and Social Research (University of Trento) and an associate member of the Department of Sociology/Nuffield (University of Oxford). He has taught extensively on social network analysis, including network theory, social network methods, and social network analysis applied to organizational, educational, and political research. His research focuses on both methodological questions and applications of social network analysis. Filip has published in leading social science journals, including the journal Social Networks. He also co-edited two special issues in the journal Social Networks: one on two-mode networks (2013) and another on negative and signed networks (2020).

Social network analysis focuses on the relations or ties between individuals, groups, organizations, countries, or other types of nodes (Borgatti, Mehra, Brass and Labianca, 2009). For example, in organizational research there is an increasing interest in the emergence of trust, advice, conflict, competition, and leadership relations between employees, and its consequences for individual and team performance. At the inter-organizational level, we might focus on the collaboration ties and exchange of resources between organizations. In educational research, people have become increasingly interested in the emergence and consequences of friendship and bullying among schoolchildren (e.g. how ethnicity, gender and personal characteristics might impact the formation network ties, and how these might affect pupils’ delinquency and smoking). In political science, we might be interested in the emergence of protest events, or conflict among nations. We might also be interested in specific ideas being propagated through social media, tracing links among criminals, the emergence of social support and its consequences for well-being and health, food-sharing networks in First Nations communities, positive and negative ties among politicians, or collaboration among NGOs.

All registrations are processed on a first-come, first-served basis. Registrations are limited, and the workshops will fill up quickly, so register today!

REGISTER HERE

Workshop Fees

A discounted fee is available for participants from academic institutions, and for participants who register in both workshops.

 

Fee

Workshop 1. Applying Social Network Analysis with R

Fee

Workshop 2. Statistical Analysis for Social Network Data

Fee

Workshop 1 and Workshop 2

Academic

500 CAD

850 CAD

1,200 CAD

Non-Academic

700 CAD

1,200 CAD

1,700 CAD

Cancellation and Refund Policies

Cancellations made prior to August 25, 2020: Full refund minus an administrative charge of 50 CAD

Cancellations made between August 25 and September 2, 2020: A 50% percent refund

Cancellations made after September 2, 2020: No refund

Overview

This 3-day workshop offers a practical workshop on how to analyze social network data with R. The workshop focuses both on how to use different R packages (such as igraph and sna) to perform social network analysis, and on how to interpret the results in specific research contexts. After discussing how to import one-mode and two-mode network data into R, we will explore different ways to visualize these networks with R. We then go on to focus on two major levels of analysis: (1) the position of individual nodes in a network, and (2) the network structure as a whole.

The first part provides an overview and discussion of different measures of centrality, such as closeness centrality, betweenness centrality, structural holes and Bonacich Beta centrality (Freeman, 1979; Burt, 1992; Borgatti, 2005; Agneessens, Borgatti and Everett, 2017), as well as centrality measures for valued networks (Opsahl, Agneessens, Skvoretz, 2010). We will discuss their substantive meaning in different contexts, and show how these can be calculated with R. This is followed by an overview of other measures of position, such as the resourcefulness and range of nodes (Burt, 1983).

In the second part of the workshop, we discuss basic structural aspects of the whole network, such as density, centralization, reciprocity, triadic effects, homophily. In addition to this, we will cover more advanced topics, such as structural and regular equivalence, core-periphery structures (Borgatti and Everett, 1999) and hierarchies (Everett and Krackhardt, 2012). Finally, we discuss ways to detect subgroups and focus on the analysis of two-mode networks (Everett and Borgatti, 2013; Agneessens and Everett, 2013).

Dates

Wednesday, September 2, 2020 to Friday, September 4, 2020

Time

9:00 AM to 12:30 PM (CST) each day

Instructor

Filip Agneessens Associate Professor in the Department of Sociology and Social Research (University of Trento) and an associate member of the Department of Sociology/Nuffield (University of Oxford)

Location

Interactive online workshop (platform will be communicated by instructor one week prior to the event)

Prerequisites

No prior knowledge of R is needed for the course

Core topics day 1: Visualizations and basic descriptive network measures
  • Importing network data in R
  • Visualizing social networks
  • Density, reciprocity, triad census
  • Degree centrality
Core topics day 2: Measures of position
  • Closeness centrality
  • Betweenness centrality
  • Bonacich beta centrality
  • Resourcefulness and range
Core topics day 3: Whole network measures
  • Centralization and core-periphery structures
  • Subgroups
  • Equivalence
  • Two-mode networks

References

Agneessens, F., Everett, M. 2013. Introduction to the special issue on advanced two-mode networks. Social Networks 35, 145-278. 

Agneessens, F., Borgatti, S.P., Everett, M.G. 2017. Geodesic based centrality: Unifying the local and the global. Social Networks 49, 12-26. 

Borgatti, S.P. 2005. Centrality and network flow. Social Networks 27, 55-71. 

Borgatti, S.P., Everett , M.G. 1999. Models of Core/Periphery Structures. Social Networks 21, 375-395. 

Borgatti, S.P., Mehra, A., Brass, D.J., Labianca, G. 2009. Network analysis in the social sciences. Science 323, 892-895.  

Burt, R.S. 1983. Range. In: Burt & Minor (Eds.) Applied Network Analysis. Beverly Hills: Sage, 176-194. 

Burt, R.S. 1992. Structural Holes: The Social Structure of Competition. Harvard University Press. 

Everett, M.G., Borgatti, S.P. 2013. The dual-projection approach for two-mode networks. Social Networks 34, 204-210. 

Everett, M., Krackhardt, D. 2012. A second look at Krackhardt's graph theoretical dimensions of informal organizations. Social Networks 34, 159-163. 

Freeman, L.C. 1979. Centrality in social networks: conceptual clarification. Social Networks 1: 215-239. 

McPherson, M., Smith-Lovin, L., Cook, J.M. 2001. Birds of a feather. Annual Review of Sociology 27, 415-444. 

Opsahl, T., Agneessens, F., Skvoretz, J. 2010. Node centrality in weighted networks: Generalizing degree and shortest paths. Social Networks 32, 245-251.

Overview

This 5-day workshop focuses on different statistical tests for social network analysis, with special attention being paid to exponential random graph models (ERGM) and stochastic actor oriented models (SAOM) using RSiena.

Statistical network analysis can be used to explain the formation of ties between nodes by focusing on both structure and nodal attributes. ERGM (Robins, Pattison, Kalish and Lusher, 2007; Lusher, Koskinen and Robins, 2013) are particularly fitted to answer questions about why ties are present in specific places in the network. For example, using ERGM we could study whether friendships are more likely to form among students if they have similar interests, or grades. We would also be able to understand whether employees in organizations are more likely to trust each other if they share an office, or are similar regarding personality.

Analyzing longitudinal data with SAOM/RSiena (Snijders, van de Bunt, Steglich, 2013), we are able to understand the process of network emergence and its consequences. With this approach we study whether smokers tend to connect to other smokers, or if employees influence each other in organizational commitment. We can also study whether friendship among scientists leads to collaboration on papers, or whether collaborative behaviour leads to the emergence of friendship.

Dates

Monday, September 7, 2020 to Friday, September 11, 2020

Time

9:00 AM to 12:30 PM (CST) each day

Instructor

Filip Agneessens Associate Professor in the Department of Sociology and Social Research (University of Trento) and an associate member of the Department of Sociology/Nuffield (University of Oxford)

Location

Interactive online workshop (platform will be communicated by instructor one week prior to the event)

Prerequisites

This workshop will make use of MPNet for analysing ERGMs, and the RSiena package in R to run SAOMs. No prior knowledge of R is needed for the course, but a basic knowledge of SNA is advisable.

Core topics day 1: Statistical analysis and ERGM
  • Principles of statistical analysis (Agneessens, 2020)
  • Introduction to ERGM (Robins et al., 2007)
Core topics day 2: ERGM principles
  • Modelling ERGMs and interpretation (Lusher et al., 2013)
Core topics day 3: ERGM advanced topics
  • Interpretation and goodness of fit
  • Advanced topics, including two-mode networks
Core topics day 4: SAOM/SIENA models
  • Principles of SAOM (Snijders et al., 2010)
  • Modelling SAOM and interpretation of structural effects and nodal attributes
Core topics day 5: Co-evolutions of SAOM/SIENA models
  • Modelling co-evolution models, and interpretation of social influence and selection effects
  • Relational event models (principles)

References

Agneessens, F. 2020. Dyadic, nodal and group-level approaches to study the antecedents and consequences of networks: Which social network models to use and when? In The Oxford Handbook of Social Networks, edited by Ryan Light and James Moody. Oxford University Press.

Robins, G., Pattison, P., Kalish, Y., and Lusher, D.. 2007. On exponential random graph models for cross-sectional analysis of complete networks: An introduction to exponential random graph (p*) models for social networks. Social Networks, 29, 173-191.

Lusher, D., Koskinen, J., and Robins, G. (eds.) 2013. Exponential Random Graph Models for Social Networks. Structural Analysis in the Social Sciences. New York: Cambridge University Press.

Snijders, T.A.B., van de Bunt, G., and Steglich, C.. 2010. Introduction to stochastic actor-based models for network dynamics. Social Networks, 32, 44-60.

Filip Agneessens is an Associate Professor in the Department of Sociology and Social Research (University of Trento) and an associate member of the Department of Sociology/Nuffield (University of Oxford).

His research focuses on both methodological questions and applications of social network analysis. He has published in leading social science journals, including the journal Social Networks. He also co-edited two special issues in the journal Social Networks: one on two-mode networks (2013) and another on negative and signed networks (2020). Besides methodological questions, his research focuses on how network relations come about, and how they subsequently impact individual and group outcomes (such as performance, well-being, etc.). He has also worked on mathematical models for analyzing complete networks and on typologies for social support (ego-networks).

He has taught extensively on social network analysis, including network theory, social network methods and social network analysis applied to organizational, educational and political research.

NVivo Training

The SSRL offers regularly-scheduled training in NVivo 12 Pro for Windows, led by SSRL Qualitative Research Manager and Specialist and NVivo Certified Expert, Rachel Tang.

The two-hour introductory NVivo 12 Pro for Windows training session utilizes a hands-on approach with a sample data set. The instructor introduces users to working with data in NVivo (importing, opening, and organizing various file types), guides learners through exploring and coding data in order to analyze for themes, and demonstrates novel program features. This workshop is most helpful for beginners or those who would like a refresher on how to manage and categorize data in order to enhance their thematic analysis skills and gain overall insight as to how NVivo for Windows is used in qualitative research.

Registration is open to all University of Saskatchewan faculty, students and staff, and employees of government, non-governmental organizations and community-based organizations. Registrations are processed on a first-come, first-served basis.

REGISTER HERE

Cost

$50.00 plus GST. A cancellation fee of $20.00 plus GST is assessed if canceled at least one week prior to the session. Refunds will not be issued if canceled within one week of the session.

Upcoming Sessions

  • Wednesday, September 2, 2020, 9:30 AM - 11:30 AM

PLEASE NOTE: Due to the ongoing COVID-19 pandemic, NVivo 12 Pro for Windows training will be offered exclusively online using WebEx. Details will be provided to you upon registration. It is not mandatory, but highly recommended that participants obtain an NVivo 12 Pro for Windows license in order to participate in the session in a 'hands-on' manner.

For more information about the training or obtaining an NVivo license, please contact the course instructor at rachel.tang@usask.ca or (306) 966-6319.

SSRL YouTube Channel

The SSRL YouTube Channel features new and archived videos from select SSRL workshops, lectures and events.