A Multidimensional Framework to Uncover Insights of Group Performance and Outcomes in Competitive Environments With a Case Study of FIFA World Cups

Abstract

Research on the performance of groups in competitive environments has traditionally focused on studying context-specific or collaboration factors without considering a multidimensional systems view integrating both. Additionally, there is limited research considering the co-dependence between the performance of a group and its adversaries. This paper proposes a framework to address these limitations by incorporating context-specific, network-based, and individual attributes to identify patterns and attributes of successful (and unsuccessful) groups. The framework provides a method to characterize performance patterns by searching for the dominant attributes that distinguish one pattern from another - relevant for decision-makers when dealing with many features. This analysis finds the different group behavior, both internal to the group and external, based on competition. The approach also identifies winning attributes through a machine-learning classification model. These factors allow differentiating a successful group and weighting context-specific network and opponent attributes. The framework is complemented with a visualization component illustrating competition with context-specific and network attributes at the player level. A case study is presented with data from FIFA World Cups in 2014 and 2018 to demonstrate the applicability of the proposed framework.