This paper investigates whether there is a relationship between local political culture and use of local media; especially the internet. The investigation is done on the basis of a case study conducted in the Danish municipality of Odder between 2003 and 2006. The case study is a very thorough investigation of local media, local politics and local democracy. In connection with the case study a wealth of qualitative as well as quantitative data were collected, among these two surveys conducted in 2003 and 2005. On the basis of a triangulation between theoretical considerations concerning the differences between a representative democratic model of political governance and a network model of governance, system politics and everyday politics, and the results of 20 in-depth citizen interviews, four (ideal-) types of citizen roles are constructed: the voter, the spectator, the project maker and the association member. These four citizen types, which represent four different local political cultures, are then sought identified in the data from the surveys, and the use of local media for all four types is mapped. It is the basic hypothesis of the paper that the patterns of media use of the different citizen types are clearly distinct, meaning that differences in local political culture plays an important role in explaining variations in the use of local media for political purposes.

                      The statistical analyses done shows firstly that the four citizen types use local media differently for political purposes, and that these differences are in accordance with the assumptions made concerning the characteristics of each of the roles. Thus, the basic hypothesis in our investigation is confirmed. Secondly, the analyses showed that the citizen role model was at least as good a predictor of variations in use of local media as education, which is often considered to be a main explanatory variable concerning such variations. Thirdly, the citizen role model was tested against a cluster model designed to maximize differences in political culture between clusters. Even though this model also turned out to be quite apt in explaining variation in local media use, the citizen role model was at least as good as this model in explaining these variations. Thus, our model of the four citizen roles/four different local cultures seems to be very robust when it comes to explaining variation in political uses of local media.