Find duplicates

PL/pgSQL

SELECT unique_key, COUNT(unique_key) AS NumOccurrences
FROM table
GROUP BY uprn
HAVING ( COUNT(unique_key) > 1 );