In this paper, we compare two different representations of Framework Programs as affiliation network: “One-mode networks”‘ and “Two-mode networks”‘. The aim of this article is to show that the choice of the representation has an impact on the analysis of the networks and on the results of the analysis. In order to support our proposals, we present two forms of representation and different indicators used in the analysis. We study the network of the 6th Framework Program using the two forms of representation. In particular, we show that the identification of the central nodes is sensitive to the chosen representation. Furthermore, the nodes forming the core of the network vary according to the representation. These differences of results are important as they can influence innovation policies.