Can a hypothesis test yield a statistically significant (say p<=0.05), yet practically meaningless result? Give an example of how this might happen.