Week 10: Real Data Is Difficult

Today’s Question

Why is real data difficult before modeling begins?

Common Problems

• Multiple tables
• Inconsistent labels
• Text fields
• Ordered categories
• Dates and times

Join Checklist

• What are the keys?
• Are keys unique?
• Are any rows unmatched?
• Did the join duplicate observations?

AI Check

A join that runs is not necessarily a join that means what you think it means.