posted by stat on .
A statistician analyzing a randomized controlled experiment has tested
Null: The treatment does nothing. Alternative: The treatment does something.
using a 4% cutoff for P-values. The P-value of the test turns out to be about 1.8%.
The conclusion of the test is
-The treatment does nothing.
- The treatment does something.
“P=1.8% means that there is only about a 1.8% chance that the treatment does nothing.”
The quoted statement is