I measured it directly.
i can't remember the name of it, but it worked like this:
there was a tube with a near vacuum with a laser at the bottom, and at the top, there was a door that would drop a ball. It would track the time until the ball broke the laser. then, using the kinematic equations, you could calculate the acceleration, whitch would be the gravitional acceleration.
for refrence
x= x(naught)+ v(naught)t + .5at^2
with x and v(naught) being zero and x(naught) being the distance the ball fell and t being the time it took.
naturally, I preformed multiple tests (about 40 I think) and took the average