Springback prediction continues to play a special role in the process design of sheet metal forming processes. During the forming process, elastic energy is stored in the sheet metal, which is released after the tool is opened. The dimensional accuracy and the function of the component in particular can be negatively affected by this. For this reason, the springback is predicted by means of numerical modeling in order to improve the process design. The simulation accuracy depends primarily on the used material model to map the stress-dependent material behavior and the quality of the experimental input data. High-strength aluminum alloys, which have a distinctive springback behavior, have a slightly pronounced tensile-compressive asymmetry. Since tensile and compressive stresses often occur in sheet metal forming processes, the consideration of this effect can lead to an improvement in springback prediction. Besides modeling the yield locus curve for describing the yield strength in the plane stress area, the yield curve data is also relevant for determining the strain hardening behavior. The work hardening is usually characterized by the tensile test or the hydraulic bulge test, which is used to describe the material behavior in the tensile stress state. A promising experimental method for the analysis of the material in the uniaxial compressive stress state is the sheet metal upsetting test. In addition to a local characterization of sheet metal materials, this test also enables the investigation of the material behavior at significantly higher plastic strains than in the tensile test. Thus, in this contribution it is examined to what extent the sheet metal upsetting test is suitable for improving the springback prediction quality of material models in sheet metal forming processes.