12th Winter School and Master Class on Longitudinal Social Network Analysis

the art network
© BarabásiLab (A.-L. Barabási, S.P. Fraiberger, A. Grishchenko, M. Resch, C. Riedl, und R. Sinatra)

12th Winter School and Master Class on Longitudinal Social Network Analysis takes place 8-11th February, 2022.

The combined event is organised at the Institute for Analytical Sociology by Christian Steglich, Anastasia Menshikova and Madelene Töpfer. The event is to be held fully online. 

Programme

The Winter School will consist of an alternation of lectures, Q&A sessions, and practical work on assignments. The Master Class will consist of an in-depth discussion of submitted papers, and consultations devoted to the analysis of participants’ data sets.

It is assumed that the participants have a good basic understanding of statistical methods, including in particular logistic regression; a good understanding of the basics of social network analysis (e.g., the textbook by Borgatti, Everett, and Johnson); and a good working knowledge of R. We have prepared a separate page with preparation material, which you can access here.

 

Zoom

The course will take place via zoom; the zoom address will be distributed to registered participants on Monday 7 February 2022 in the morning hours.

Course Material

The course material will be available at this shared drive location (please click here). It will be updated throughout the course week, so be careful to update material that you may have downloaded, and be careful with printing.

Schedule

Tuesday

The first course day’s topic is the introduction of the stochastic actor-oriented model (‘SAOM’) for network dynamics.

Morning sessions

9:00-10:30

  • Short introduction round of participants
  • Types of research questions, model assumptions
  • Model components, data structures

11:00-12:30

  • Estimation, simulation and convergence
  • Architecture of the RSiena software
  • Lab exercise: The dynamics of collegial advice seeking

Afternoon sessions

14:00-15:00

  • Hypothesis testing: Wald type and score type tests
  • Goodness of fit assessment

15:30-16:30

  • Model selection
  • Lab exercise: Stepwise model construction

16:30-17:00

  • Time for open discussion

Wednesday

The second day’s topic is the extension of the stochastic actor-oriented model to the co-evolution of multiple dependent variables.

Morning sessions

9:00-10:30

  • Dynamics of co-evolution between network and behaviour
  • Network autocorrelation, conjugate explanations (selection and influence)
  • Lab exercise: The dynamics of friendship and alcohol consumption

11:00-12:30

  • Lab exercise continued
  • Model-based decomposition of autocorrelation into selection and influence
  • Lab exercise: Evaluation of simulated data on custom dimensions

Afternoon sessions

14:00-15:00

  • Two-mode networks, multiplex networks, and multilevel networks
  • Lab exercise: a multidimensional analysis

15:30-16:30

  • Lab exercise continued
  • Writing it up, laying it down: What is good documentation practice?

16:30-17:00

  • Time for open discussion

Thursday

The last day of the Winter School is devoted to advanced topics of your choice, we will make a vote and address the following topics in their order of popularity, until time is up. That will be at 13:00, a bit later than the morning sessions on the other two days. In the afternoon, you are kindly invited to a seminar presentation in the IAS seminar series – and we close the Winter school part in a final wrap-up session.

Morning sessions

9:00-10:00, 10:30-11:30, and 12:00-13:00

Sessions to address a selection of these topics:

  • Data from multiple groups
  • Dealing with absent (e.g., missing) data
  • Diffusion models
  • Marginal effects
  • Non-standard estimation algorithms
  • Ordinal networks
  • Undirected networks

Afternoon sessions

14:30-16:00

  • IAS seminar talk by our fellow course participant Elisa Bienenstock. More information (and the Zoom link) will be distributed beforehand.

16:30-17:00

  • Time for open discussion, and farewell to those not attending the Master Class.

Friday

The Master Class day will be devoted to the discussion of work-in-progress by the participants; this can be in an advanced stage (e.g., results of preliminary analyses) or still early (e.g., data collection, analysis strategy). A detailed schedule of the Friday will be distributed at the beginning of the course week.

Morning sessions

9:00-10:00, 10:30-11:30, and 12:00-13:00

Afternoon sessions

14:30-15:30 and 16:00-17:00

About schedule

The following is a rough outline of how the course topics will be distributed over the week. Afternoon sessions are shorter.

There will be longer breaks of 30 minutes between sessions and short breaks around each full and each half hour within sessions and much opportunity for asking questions. 

Preparation

Background

Winter School, February 8-10

The Winter School introduces participants to the analysis of longitudinal, group-centered network data by way of stochastic, actor-based models (Snijders, van de Bunt & Steglich, 2010), and to the analysis of peer influence processes taking place in such dynamically changing networks (Steglich, Snijders & Pearson, 2010). The objective of the Winter School is that course participants develop an understanding of the models, familiarise themselves with the use of the RSiena software for model estimation, and learn how to tell a good model specification from a bad one. The Winter School will be taught by Christian Steglich with the support of Anastasia Menshikova. This is the second time (after 2017) the Winter School will take place in Norrköping. Previous editions took place at the University of Groningen (2005-2015, 2019) ETH Zürich (2016), and online (2021).

Master Class, February 11

The Master Class on Longitudinal Social Network Analysis is an extra day of deepening your experience with longitudinal network modelling. There are two ways how you can make use of this opportunity. On the one hand, we reservetime forconsultations about getting your data analysis started, if you bring own data with you and have not yet started analyzing it. On the other hand, it is possible todiscuss your work in progress. If you have already started analyzing your data, this is an opportunity to get expert feedback on your preliminary work. Invited discussant will be Tom Snijders and (depending on paper content) maybe others; they might “zoom in” for participating.

The procedure for the Master Class is the following. Participants that would like to actively participate should submit an abstract in which they explain their analysis (or analysis plans)This abstract should be handed in together with the main application for the Winter School. Within one week after the application deadline, authors will be informed about acceptance for the Master Class. For accepted preliminary work, we ask you to submit papers of no more than 12 pages, accompanied by an R-script and (if possible) data, to enable reproducibility of the analysis, by January 21. Participation in the Master Class is optional. If you do not wish to participate in the consultations or paper discussions actively, you can still join the Master Class event as a listener and contribute to the discussions.

Prerequisites and Preparation

Prerequisites for participation are familiarity with basic social network analysis, some knowledge of intermediate statistics, and basic familiarity with the R statistical software environment. We expect the participants to bring their own laptops and install the required software beforehand

 

We especially invite researchers who are in the process of collecting or analysing their own longitudinal data sets to participate in the Master Class. For participants without own data, several sample data sets will be made available. 

 

If you have questions regarding the contents of the classes, please, address them to Christian Stieglich. The questions regarding the application procedure and the event should be sent to Anastasia Menshikova. Madelene Topfer will be available to discuss payments.

 

For an overview of earlier Winter Schools on Longitudinal Social Network Analysis, click here

Contact

Visitor information