All the methods used for the in vitro measurement of the SPF, the universal indicator of sunscreens efficiency, rely on a spectrophotometric analysis. What can vary about the experimental protocol used is mainly the substrate and the type of spectrophotometer chosen. We decided to work with polymethylmetacrylate plates that we analyzed using two spectrophotometers equipped with integrating spheres, the UV1000S and the UV2000 apparatus. Two marketed products were such tested, after spreading 2 mg/cm2 on the plates, using one apparatus after another. We applied a non-parametric Wilcoxon test for paired data to the measures realized on 10 plates (as we systematically used the 2 apparatus), in order to compare the series of measures obtained with the two machines. This way, we were able to show a significant difference between the SPF values respectively obtained with the UV1000S and the UV2000 spectrophotometers. This difference could be explained by the decrease of the stray light in the case of the UV2000 apparatus.