Guided Notes
Cleaning Wrong Formats, Wrong Data & Duplicates
Converting formats with to_datetime and to_numeric, fixing impossible values, and removing duplicates.
Data in the wrong format
Slide 1 / 5A column is only useful if every value has the right type. Common problems:
• Dates stored as text, some written differently ('2020/12/20' vs '20201226'). • Numbers stored as text, with symbols ('$1,200'). • Categories spelled inconsistently ('IT', 'it', 'I.T.').
Convert the whole column to the proper type; anything that can't be converted becomes missing, and you handle it like an empty cell.