SQL: Count duplicate records
SELECT YourColumn, COUNT(*) TotalCount FROM YourTable GROUP BY YourColumn HAVING COUNT(*) > 1 ORDER BY COUNT(*) DESC