Exponential random graph models: explaining strategic patterns of collaboration between artists in the music industry with data from Spotify

Given the abundance of online data, the exponential random graph models are helpful tools that enable researchers to answer causal inference questions considering cross-sectional network data. This chapter explains (a) what an ERGM is, (b) what kind of research this class of model can help with, (c) how to conduct the analysis, and (d) the type of conclusions that can be drawn. A case study focusing on patterns of collaborations between Rock musicians employing network data from Spotify is presented to show evidence about how to conduct this type of research. The empirical results from the analysis of the Spotify sample provide statistical evidence for the pattern of collaboration between Rock musicians. Some musicians decide not to collaborate; others collaborate only with another musician or band. Besides, musicians who employ the strategy to have more than one collaborator produce music with musicians or bands who already work with their collaborators exploiting the existing network of relationships. This chapter should provide ideas and prompt relevant research in the Social Computing field.