Currently ethylene vinyl acetate copolymers (EVA) with peroxide cross-linkers are used for the embedding of photovoltaic (PV) cells. During the lamination of PV-modules the EVA embedding material is chemically cross-linked at about 140°C. The degree and quality of cross-linking has a significant impact on the service properties and the long-term performance of PV-modules. The degree of cross-linking of EVA is currently examined by a time-consuming wet-chemical method (gel content determination). Up to now physical characterization methods have not been investigated and applied systematically. Hence, the main objective of this diploma thesis is the evaluation of various physical methods for the determination of the degree of cross-linking of solar grade EVA. Four different commercial EVA embedding films (2 Standard-Cure and 2 Fast-Cure grades) were selected for the investigations. The uncured EVA films were laminated with two release films and a glass pane and cured in a PV-module laminator. To obtain different degrees of cross-linking the lamination time was varied systematically. The EVA-films were characterized by various analytical (Differential Scanning Calorimetry (DSC), Dynamical Mechanical Analysis (DMA)) and technological (tensile test) methods. To evaluate the physical methods used, the measurement parameters were varied. The thermoanalytical investigations (DSC) under nitrogen atmosphere revealed a clear correlation between post-cross-linking enthalpy and lamination time. In contrast, no consistent trends were obtained in ambient air atmosphere. With regards to DMA the shear mode was much more sensitive compared to the tensile mode. As relevant values for the degree of cross-linking the maximum of the loss factor and the shear modulus in the melting regime were evaluated. The tensile tests at room temperature exhibited significant lamination time dependent differences in the stress-strain behaviour at high displacement levels. As sensitive value for the degree of cross-linking the elastic modulus at a strain of 500% was deduced.