Low crested structures are primarily designed as a coastal defence mechanism, making the prediction of the energy in the lee of the structure of utmost importance. As wave transmission is a measure for the energy that passes the breakwater, experiments on oblique wave transmission have previously been performed by the European project DELOS (Environmental Design of Low Crested Coastal Defence Structures) and analysed further by Van der Meer et al. From these experiments, it is unclear whether it is the roughness or the permeability of the core that determines the behaviour of the structure. Therefore, the objective of this study is to improve the understanding of oblique wave transmission through rough impermeable rubble mound submerged breakwaters by means of a 3D physical model. An experimental study is conducted based on the same set-up of DELOS but the permeability of the core of the rough structure will be varied. Finally, by comparing the data of this study with the data of DELOS and the formulations of Van der Meer et al., insight should be gained on the matter. In total, four structures are tested using irregular long crested waves; ranging from a fully permeable to a fully impermeable rough rubble mound breakwater. In order to test for oblique waves, the model is rotated progressively by 15°, ranging from 0° to 60°. Secondary effects due to the physical constraints of the wave basin are minimised by placing an additional mound of rubble between the model and the edge of the basin. This prevents a large scale circulation pattern from occurring and reduces wave diffraction effects. The directional spectral analysis software DIWASP is used to calculate the variance density spectra of the data. The wave climate is estimated with the IMLM method (Iterated Maximum Likelihood Method) because it is the most suited to portray the narrow directional spread. The aim of this study is to analyse the influence of the incident wave direction βi on the transmitted wave direction βt, the transmission coefficient Kt and the spectral changes of the transmitted spectrum. An analysis of the data shows that for rough structures there is no significant change in wave direction. The incident wave direction is approximately equal to the transmitted wave direction (βt=0.94βi for 0° ≤ βi ≤ 60°). The data of this study also show a slight increase in the transmission coefficient with increasing incident wave angle. However, when considering a combined data set, which includes the data of this study and the rough permeable data of DELOS, the data show that oblique wave attack has a negligible influence on the transmission coefficient for rough structures. The spectral changes of this study support the model proposed by Van der Meer et al. Finally, it is concluded that it is the roughness of the structure rather than the permeability of the core that determines the behaviour of the breakwater with respect to the incident wave direction.