Data Cleaning Exercises

Module 5

Categories and mappings

Approved dictionaries, unmatched values, controlled vocabularies and reversible mappings.
  1. Map categories with an approved dictionary

    Normalize category spellings, map them with a reviewed dictionary, and quarantine values the dictionary doesn't cover. · 9 min

  2. Unmapped values go to exceptions

    Handle values outside an approved dictionary without guessing. · 7 min

  3. Mapping tables as data

    Keep category mappings in a reviewed table shared by every tool. · 6 min

  4. Reviewing a mapping change

    Change a category mapping safely and measure its effect before releasing it. · 6 min

Module 5 quiz

Practice questions: choose an answer and check it to see the explanation. Nothing here is timed or graded.

  1. 1. Why normalize case and spacing before mapping?
  2. 2. A channel value "emial" is not in the approved dictionary. What should happen?
  3. 3. Why should a category mapping be stored as data (a table) rather than as scattered IF statements?
  4. 4. You lowercase and trim channel, then map using a dictionary whose keys are "Email" and "Phone". What happens to "email"?
  5. 5. What makes a mapping reversible?
  6. 6. Two source spellings could map to two different approved categories. What should you do?
  7. 7. A mapping has 7 entries; the data has 9 distinct normalized values. How many values end up unmapped at most?