Mathematical methods to statistically compare the relative accuracy of physiological monitoring devices to reference measurements are scarce. We developed such a method to compare any number of devices to each other as well as to reference values. This new statistical comparative procedure uses a combination of different known mathematical processes. The method allows the separation of the error of the devices into two components, the first, being systematic, can be easily amended, while the second, inherent to the technical limitations of the measuring device as well as to the physical properties of the explored phenomenon, cannot be corrected and will thus be used as the comparative criterion for statistics. That method, which is easy to implement should ease comparisons of the accuracy of any number of devices as well as comparison of other sets of physiological values, by giving an easy access to statistical results.