To date, not much is known about how the functions of music relate to music preference. This article examines the basic hypothesis that the strength of preference for a given kind of music depends on the degree to which that kind of music serves the needs of the listener; that is, how well the respective functions of music are fulfilled. Study 1, a pilot study, identified the best-known musical styles of the participants, yielding 25 styles that were known by at least 10 percent of them. Study 2 used these 25 styles and found that rock, pop and classical music were liked most. A factor analysis yielded six distinct dimensions of music preference. People showed great variation in the strength of preference for their favourite music. This is explained by the impact of different functions of music. The potential of music to express people's identity and values and to bring them together was most closely related to the strength of preference. However, the reasons for liking a particular style are not congruent with the functions that people ascribe to their favourite music in general. A theoretical model of the development of music preferences is suggested. (Contains 6 tables and 3 figures.)