The effects of Internet use on the academic performance of university students is usually analyzed from a deterministic perspective that gives this technology an immanently positive or negative role. In this paper we consider the problem from an alternative point of view. We analyze the actual Internet uses the students engage in and the differences in these according to the students’ characteristics. We aim to explain when spending more time on the Internet has a positive and a negative effect on academic performance and explain why usually a high usage time of internet is correlated with a lower academic achievement.