This workshop is about analysing mobility networks, that is, networks in which nodes represent locations and ties are individuals that are mobile between these locations. Examples of mobility networks include migration of individuals between countries and mobility of workers between organisations. Mobility networks as understood here are directed and weighted. The workshop teaches a statistical method to analyze such data, which is introduced in “Block, P., Stadtfeld, C., & Robins, G. (2022). A statistical model for the analysis of mobility tables as weighted networks with an application to faculty hiring networks. Social Networks, 68, 264-278.”. The method is implemented in MoNAn, a package of the statistical system R. The workshop will demonstrate the basics of using MoNAn. Attention will be paid to the underlying statistical methodology, to examples, and to the use of the software.

The goal of this method is to model endogenous (network) patterns in mobility networks, such as concentration, reciprocation, and triadic clustering. The prevalence of these endogenous structure can be modelled alongside classical predictors of mobility that concern attributes of individuals and locations (i.e., “controlling for” these predictors). As such, it is in the spirit of ERGMs but applies to mobility data. Technically, the presented model represents an extension of classical log-linear models applied to mobility tables.

The first part of the workshop will focus on the intuitive understanding of the model and operation of the software. The second part will present a deeper treatment of the statistical model and detailed introduction into some advanced features of the software, for example, goodness of fit, or advanced model specification.

A basic introduction of the software and pointers to further material is provided on the MoNAn github page (github.com/stocnet/MoNAn).

Prerequisites:

Course participants should have a basic understanding of model-based statistical inference (say, logistic regression), some prior knowledge of social networks, and should have had some basic exposure to the R statistical software environment. They are expected to bring their own laptop to the course (Windows, Mac or Linux), with the R statistical software environment and the MoNAn package pre-installed. Participants for whom R is new are requested to learn the basics of R before the workshop: how to run R and how to give basic R commands. This is to reduce the amount of new material to digest at the workshop itself. Further instructions will be given before the conference starts.

Organiser:

Per Block, University of Zurich, Department of Sociology.

email: per.block@uzh.ch

Workshop length:

6 hours.

Max Participants:

30