Fábio Generoso Vieira - Roger Leenders - Joris Mulder
Modeling heterogeneity of social interaction behavior across different temporal networks
My PhD project is focused on developing multilevel models for dynamic social network data. In this framework, observations are nested in groups, each representing an independent network containing actors sampled from the population under study. So far, we have developed multivariate network models to analyze these data by estimating the covariate effects in the population and conducting formal hypothesis testing methods to examine both significances of effects and to investigate scientific hypotheses.
Software package
REMSTATS
The R software package remstats was developed for efficient computation of statistics for analyzing relational event history data with relational event models (REMs). The software package includes a wide range of endogenous and exogenous statistics for a tie-oriented REM (Butts, 2008) or actor-oriented REM (DyNAM, Stadtfeld & Block, 2017). It can handle directed and undirected events and allows for REMs that consider event types in the dependent variable. The resulting object can be used with the R software packages remstimate or relevent (Butts, 2008) to obtain estimates for the model parameters. The package can be retrieved from GitHub.
Applications
Social interaction and student rebellion
So far, I have worked with the classroom data as an empirical application to my research. These data were collected by Daniel McFarland during a study to investigate student rebellion in the classroom. The data contain interactions amongst high-school students in the US, nested in classrooms. I have used these data to illustrate our model and provide some insights about social interaction behavior in the classroom. Among our findings, we show confirmation of the unique role of the teacher in conducting the flow of communication in the classroom, and we also present results pointing to the significance of networks effects, such as outgoingness and popularity.
Papers
Bayesian mixed-effect models for independent dynamic social network data.
Meta-analytic approximations for relational event models.
Empirical Bayes factor for testing random-effect structures.