Data from a Swedish PISA-sample were used (1) to identify a digital reading factor, (2) to investigate gender differences in this factor (if found), and (3) to explore how computer game playing might relate to digital reading performance and gender. The analyses were conducted with structural equation modeling techniques. In addition to an overall reading factor, the hypothesized digital reading factor was identified. When the overall reading performance was taken into account, a relative difference in favor of the boys for digital reading was indicated. This effect was mediated by a game-playing factor comprising the amount of time spent on playing computer games. Thus, the boys' better performance in digital reading was explained by the computer game-playing factor.