I would like to get frequent element of the column in SQL.
For instance,
Assume we have the below column,
Jack
Jack
Jack
Jeremy
Katie
Katie
John
I would like to get Jack.
Furthermore in below case,
Jack
Jack
Jack
Jeremy
Katie
Katie
Katie
John
I would like to get Jack and Katie.
What I have tried?
GROUP BY COLUMN ORDER BY COUNT(*) DESC LIMIT 1;
It works in first case, but since we are limiting it by 1, second case fails and only returns Jack instead of Jack and Katie.
TL;DR I would like to get most frequent elements even if their frequency are the same.
Thanks in advance.