Why are we even talking about correlation? The study performed statistical hypothesis testing to determine that the means of the two groups are probably differ

Why are we even talking about correlation? The study performed statistical hypothesis testing to determine that the means of the two groups are probably different. Edit: and more to your point, correlation of the sample implies correlation of the population if the sample is representative. Is this the case for this particular paper? I don't know and neither does anyone else; that's why replication is the gold standard of scientific validity.