Conference Agenda

We first give a mathematical definition of multimode data this includes directed multimode data. Most 2-mode data is undirected but it is possible to have directed data, this would require both modes to have some degree of agency. We give some examples of directed two-mode data and suggest techniques for analyzing and representing such data. We examine standard methods but also look at projections and core-periphery models including the dual projection approach.

Interventions in networks is becoming an increasingly more important topic in public health, business, and public policy more widely concerned with opinion change and online networks. Yet, network research suffers a lack of viable causal frameworks and the choice of targets of interventions are typically based on heuristics. Optimal design theory has long been established as a statistical technique for designing experiments, interventions, etc, but has thus far rarely been applied to networks. A notable exception is Parker, Gilmour, and Schormans (2017), who proposed optimal designs for networks but only for a linear model, i.e, with independent (SIC!) outcomes. We explore designs for network effects and network autocorrelation models, and how these are affected by increasing network autocorrelation and social influence - whom to do treat if potential targets have the capacity to influence others? We derive the local D-optimal design for a specific illustrative example. However, this design is not unique, and highly sensitive to the strength of network autocorrelation. In terms of designs, network models suffer from similar limitations as standard non-linear models but, in addition, determination requires that the network is fixed and known. We suggest that a way forward is to consider designs that are optimal in expectation, either with respect to an a priori network model such as the ERGM or autocorrelation coefficient.

Recently, network methods have become increasingly popular for analysing various types of mobility. In these methods, the rows and columns of mobility tables (e.g., occupations) are treated as network nodes connected by mobile individuals, creating a directed, weighted mobility network. Originally, descriptive network methods like community detection have been used alongside statistical methods such as exponential random graph models (ERGMs) developed for binary networks. Recent extensions to log-linear models have further integrated network approaches to the study of mobility. These models challenge the dominant covariate-based approach to mobility by explaining mobility through endogenous network structures, such as reciprocation or clustering, that represent emergent social mechanisms. In this study, we empirically assess the necessity of modelling endogenous structures across five types of mobility: intergenerational occupational mobility, marriage mobility, residential mobility, international student mobility, and “money mobility” derived from a behavioural experiment. For each type, we compare the fit of (i) a log-linear model with one endogenous (network) parameter, (ii) a log-linear model with one measured covariate, and (iii) such a model with one latent covariate estimated to maximize the model fit. Model fit is assessed by the extent to which individuals from the same origin select the same destination, i.e., bandwagon effects in mobility. We find that the network model consistently fits the data best. Our results suggest that researchers interested in such bandwagon effects should consider using network methods. Furthermore, our study offers a method to compare different models using flexible fit criteria that relate to researchers’ modelling goals.

This presentation focuses on the analysis of multilevel networks over time. Multilevel structures are complex systems that combine different types of ties in various domains where the aim of the modelling is to reduce the structural complexity of networks and provide a meaningful interpretation. We will demonstrate a graphical representation of multilevel structures over time and conduct a positional analysis of multilevel networks with different types of ties. Algebraic methods will be used to represent the role algebra for the constructed positional system, and the presentation will also discuss challenges and future directions in analysing multilevel structures and dynamics in complex networks.

The biased net framework, originally introduced by Solomonoff and Rapoport in the 1950s, has undergone many iterations over the decades. Recent work (building on work by Skvoretz and colleagues) has proposed an approach based on a Markovian specification in which observed networks arise from a latent dynamic process driven by "activation" events (which result in tie formation or persistence) and "inhibition" events (which prevent or remove ties). Although the likelihood arising from this process is incomputable, approximate Bayesian inference can be performed using an approximate prevision strategy, in which a non-parametric least-squares learner trained on synthetic data is used to infer posterior expectations from observed graphs. While proof-of-principle work with this approach has made use of model terms based on traditional biased net concepts (e.g., parent, sibling, and double-role effects), it is also possible to specify models using terms more closely related to more modern exponential-family random graph and stochastic actor-oriented frameworks. These allow for more natural specification of covariate and dependence effects, including both activation and inhibition variants. Here, we discuss some of these new specifications, and illustrate their application to empirical cases.