Mike E wrote:
"Here is an example. If students are trying to see how well the
velocity
of a glider on a level air track stays constant, and/or to see if air
friction is observable, and their measurements show a velocity of 34.88
cm/s near the start and 34.64 cm/s near the end of the track, then they
would not want to say, "the two velocities were 35 cm/s and 35 cm/s so
the velocity stayed exactly constant.""
Ideally, you might want them to say "Since one measurement gives no
information about the uncertainty in the measurements, then I can
conclude nothing from these two numbers by themselves."
Mike goes on to say that repeated measurements give the same effect. In
this case, the answer in the report ought to be something like "on
average the carts slowed down by 0.2 cm/s +/-0.2 cm/s and I am 98%
certain that this result is significantly different from 0 cm/s".
I have very rarely seen a lab course where this sort of statistical
hypothesis testing was encouraged.