A Large-Scale, Longitudinal Study of Player Achievements in World of Warcraft
A Large-Scale, Longitudinal Study of Player Achievements in World of Warcraft Jonathan Bell, Swapneel Sheth, Gail Kaiser Department of Computer Science, Columbia University, New York, NY 10027 {jbell, swapneel, kaiser}@cs.columbia.edu ABSTRACT studies typically involve conducting user studies, collecting We present a survey of usage of the popular Massively Mul- information through online surveys. Quantitative studies, tiplayer Online Role Playing Game, World of Warcraft. By on the other hand, are typically based on more longitudi- mining publicly available data, we collected a dataset con- nal data, typically game metrics recorded by game studios sisting of the player history for approximately six million or researchers. In either case, the information gathered can characters, with partial data for another six million char- be used in many ways within game development, allowing acters. This paper focuses on player achievement data in developers to customize users' interactions with the game. particular, exposing trends in play from this highly success- For instance, work has been done to identify player's moti- ful game. From this data, we present several findings on vations for playing [20], and to correlate those motivations players' play styles. We correlate achievements with moti- with player retention [7], allowing developers to use this data vations based upon a previously-defined motivation model, to improve user retention. and then classify players based on the categories of achieve- We recognize that the quantitative study of games is a ments that they pursued. Experiments show players who developing field, with many data sources that have not been fall within each of these buckets can play differently, and fully exploited, and that there are still questions unasked.
[Show full text]