Data quality depends on fitness for a specific use. Completeness, validity, consistency, uniqueness and freshness require separate checks.
Before you begin
Use fictional or properly licensed data with a documented question, units and column definitions.
Work through the idea
State the decision or learning question first. Keep raw inputs separate from derived tables and document filters, exclusions and calculations. Check at least one result manually and record a limitation. Communicate what the evidence supports without inventing business outcomes or exposing personal information.
- Define a rule for each important field
- Separate errors from warnings
- Track unresolved exceptions
Practice task
Validate a fictional customer export with duplicate IDs, missing fields and inconsistent country codes.
Check your result
Report counts and examples without publishing real identities.
A mistake to avoid
Do not describe a dataset as accurate only because it has no empty cells.
Keep a review note
Save the inputs or scenario, your reasoning and the result you checked. Explain one assumption and one case that needs more work. Revise the exercise after feedback and distinguish what you demonstrated from what remains untested.
Explore the related learning path · Choose another worksheet