A statistical model for the analysis of mobility tables as weighted networks with an application to faculty hiring networks

Mobility of individuals between a wide variety geographic locations, social positions, or roles is frequently analysed in the sciences. In recent research, mobility has increasingly been conceptualised as network. For example, residential mobility, when move neighbourhoods city, can be understood network which are nodes that tied by counts mobile from one neighbourhood to another. Understanding allows apply concepts and methods analyst's toolbox. However, statistical modelling such weighted networks ties have individual attributes remains difficult. this article we propose model for analysis tables networks, combining properties log-linear models exponential random graph (ERGMs). When no endogenous patterns modelled, it reduces classic tables. but ignoring attributes, an ERGM tie weights denote counts. Making use special constraints offers parsimonious way deal with ties. Going beyond classical ERG modelling, proposed approach additionally incorporate characteristics represent people. The applied two cases faculty hiring networks—linking current departments members their PhD granting institution—in history computer science US.