Networks, Gold, and Code: New Findings from the Virtual World Exploratorium

Dmitri Williams · Noshir Contractor · M. Scott Poole · Jaideep Srivastava

Fri., June 12, 11:00–12:30, Browsing Library (2nd floor, West Central)

The Virtual World Exploratorium (VWE) team uses data logs, surveys, and privacy-protected network data to study the behaviors and profiles of MMO players in the game EverQuest II. By combining large-scale surveys with supercomputing analysis of game logs, the project investigates a wide range of social and computer science phenomena. In this presentation, the team will present the latest findings, focused this year around the topics of sociability, networks and economic behaviors. Five new sets of results will be presented:

• Member groups: Network analysis will demonstrate how to track groups and interactions over time, with an emphasis on the homophily of groups (how alike are the members?) and the real-world proximity of the members.
• Trust in MMOs: Large-scale survey and behavioral data will show who trusts whom, how much, and when it varies.
• Grouping behaviors and effectiveness: Game-based performance measures are combined with profile data to demonstrate how well different kinds of groups and sizes of groups perform. Differences between the incentives of the game and the reality of the players will be highlighted.
• Gold farming: Data will be presented to show the behavioral patterns and real-world locations of gold farmers.
• Economic patterns: Large-scale aggregated economic data will show, for the first time, exact values of GDP for a virtual world, as well as to illustrate the mapping principle of virtual worlds paralleling real ones.