Abstract
Nowadays, many of the top-selling video games include options to purchase loot boxes as paid virtual items. As research progressed, loot boxes have been found to have similar characteristics to gambling, and there has been an ongoing debate as to whether loot boxes can be defined as gambling. In order to better study loot boxes, psychometrically meaningful scales are necessary. The Risky Loot Box Index (RLI) was developed by Brooks and Clark, which is the most commonly used tool to assess the use of loot boxes. This study aimed to translate the original RLI into Chinese and evaluate its psychometric properties. Two samples were recruited through online gaming forums (n = 143) and offline internet cafes (n = 236). An exploratory factor analysis of the online sample yielded a one-dimensional nine-item model, with the factor focused on risky behaviors associated with loot boxes. The confirmatory factor analysis carried out on the offline sample corroborated the results obtained from the exploratory factor analysis, and the Chinese version of the RLI displays satisfactory psychometric properties. Furthermore, the Problem Gambling Severity Index (r = 0.57, P < 0.001) and the Internet Gaming Disorder Scale-Short Form (r = 0.67, P < 0.001) were found to be significantly associated with the RLI. We also found that players with high RLI scores may have higher levels of anxiety and depression, and they were more willing to spend money on loot boxes, with some spending nearly all their earnings. Interestingly, no significant correlations between age, gender, education, or income level, and the RLI were found.