They haven't.The bigger issue here is the accuracy of the program in question and the results. Since NIST seems to have agreed with Chandler, then any attempts to prove chandler wrong must also prove NIST wrong.
NIST did two data fits. The linear regression is a way of finding the average acceleration, and both Chandler and NIST agree on what the average acceleration is for a certain period of time. That's all they agree on. However, NIST did also another curve fit, which goes over g. Chandler didn't do this fit. NIST's analysis leaves the door open to the possibility that the acceleration wasn't constant; Chandler's doesn't.
There are reasons to think that the acceleration indeed went over g for a certain period of time, therefore the average acceleration is not enough to study the behavior of the building, and thus the linear fit is not the whole story, but just what it is: an average.
Last edited:
