Shows that Special Relativity (SR) cannot explain net proper time differences (NPTDs) in such areas as GPS.
Step 1: Note that the data shows that all clocks at rest in the same frame, other things being equal (e.g., same gravitational potential), have the same clock rate (i.e., accumulate proper time at the same rate). [This is the basis for having the standard second, etc.]
Step 2: SR claims all inertial frames are equivalent. Hence, if SR claims that accelerating from being at rest in frame A to being at rest in frame B reduces the clock rate for the accelerated clock, then if SR is consistently applied, SR must also claim that accelerating from being at rest in frame B to being at rest in frame A reduces the clock rate for the accelerated clock.
Step 3: Consistent with Step 2, if we take a clock from frame A and move it to frame B and then move it back to being at rest in frame A, it will, according to a consistent application of SR, have undergone two, cumulative clock slowings. However, we know that this prediction of cumulative slowing is at odds with the empirical data of Step1.
QED
Note: As discussed elsewhere on this site, GPS clock slowing data is asymmetric whereas SR, if applied consistently, requires symmetric time dilation.