Examining the relationship between behavioural patterns, game classifications and microtransactions in digital games

Abstract

Microtransactions are a relatively new entry in the field of digital games. They refer to purchases made separate from the acquisition of a license key needed to play the game. Recently, they have come under criticism for their employment of gambling-like methodologies used to entice users into further transaction. In this study, we gathered information from the local community of users that play digital games in an attempt to observe potential engagement risks that microtransactions may entail. We specialized our data set to fit criteria relevant to the real-world scenario by observing literature on games, gambling and data science. We developed a machine learning Decision Tree algorithm to digest our data set and develop a predictive model, from which we could infer relationships between data associations. From our results, we found young students to be the most at risk of becoming engaged in these methodologies, primarily in competitive games that employ lootboxes with cosmetic customizations.