Discrepancies that are Statistically Significant

I think I understand how base rate can be clinically significant but I'm really confused how to tell when something is statistically significant. How do we establish the range for statistical significance and which scores to do you consider when comparing them to this range?