An Ego-Centered Analysis of Large-Scale Networks: The Case of the Marketing Discipline


Speaker


Abstract

Computerized databases of collaboration among academic researchers have become an important data source for the analysis of large-scale networks. This paper presents the first results of the Marketing Connectivity Project (www.mconnectivity.com) in which we document and analyzethe evolution of collaboration among marketing academic researchers going back more than 40years. Based on the ProQuest database and a software program, we document the social collaboration among researchers in dozens of the leading marketing journals, enabling us to create the ego-centered networks of active marketing researchers, and to identify, in a network framework, the most central marketing researchers. Unlike most published academic collaboration research, our analysis is not limited to a snapshot of the current situation, but is dynamic, and follows the evolution of the field over decades. It also allows us to examine the structure of, and possible ways to model, ego-centered networks in general. Similarly to many other social networks, the marketing academic community has been identified as a small-world community. The average separation among the more than 15,000 connected individuals in the 2004 dataset is low relative to network size (8.1). We find that the level of collaboration and the coherence of the marketing network increased over the last few decades, yet it is still not as strong compared to the more established scientific disciplines to which we compared marketing. An interesting finding has been the decline in the average separation over time even as the marketing network has grown considerably. This finding demonstrates how the connectivity structure, much more than network size, can drive the average separation between individuals.