Cleaning Real Data: Missing, Duplicated and Wrong · seed 1 · A4, ink-friendly. The answer key prints on its own page for grown-ups.
En anglais

Clean it and write down the rule

Computing · Data & Databases · ages 21-22
Name ______________________   Date ____________

  1. Which three moves handle missing values?

    • Drop, fill, or flag with a reason
    • Fill every blank with zero, then move on
    • Ignore blanks, since averages skip them

  2. A city column holds " Paris", "PARIS" and "paris ". Which single rule makes them one city?

    • Keep only rows that match exactly
    • Lower case and trim spaces
    • Sort the column from A to Z

  3. Every cleaning repair can change the result.

    Circle one:   True   False

  4. A column is a third missing, mostly in one region. Dropping all those rows erases the region. What is the safer move?

    • Fill each gap with the overall average and move on
    • Flag the gaps and show how filling moves the answer
    • Copy one region over the missing rows

  5. Two rows share a name but carry different id numbers. What are they?

    • Two things, so keep both rows
    • One thing, so merge them at once
    • Trash, so delete both rows

  6. How could a cleaning choice change the conclusion?

    • Dropping rows can erase one group and tilt the average
    • Cleaned data always prints in color
    • Rules make files open faster

  7. When is dropping the right call for blanks?

    • When the gaps are few and unrelated to the question
    • When you want the file to look short
    • Always, since blanks are ugly

  8. A shop appears as Main St and Main Street with the same id. Match or not?

    • Keep apart, since letters differ
    • Keep both, since two rows always mean two shops
    • Match them, since the stable id agrees
LightMySky · lightmysky.comW1-mt_qIsLXA-7EC-s1

Answer key

For grown-ups. Fold this page away before handing over the rest.

Clean it and write down the rule W1-mt_qIsLXA-7EC-s1

  1. Drop, fill, or flag with a reason · Every blank needs one of the three moves plus a stated reason.
  2. Lower case and trim spaces · One written matching rule collapses the three spellings into a single city.
  3. True · Each fix reshapes the data the answer is built from.
  4. Flag the gaps and show how filling moves the answer · Flagging keeps the gap visible and shows how the fix moves the answer.
  5. Two things, so keep both rows · A stable id never shifts like spelling, so different ids mean different things.
  6. Dropping rows can erase one group and tilt the average · Removing a group reshapes whatever average or count you report.
  7. When the gaps are few and unrelated to the question · Few unrelated gaps barely move the answer, so dropping stays defensible.
  8. Match them, since the stable id agrees · Matching spellings plus one shared id means copies in disguise.
Worksheet · LightMySky