Activity: Clean a sample dataset

• Individually or in small groups. Could also be a large group walkthrough.
• Select a small dataset that needs to be cleaned. Or, find a cleaned/structured dataset and add some issues to the dataset. Note that you have made alterations to the original dataset if you select the second option.
• Ask attendees to clean the dataset while documenting along the way. Can be done using a specialized tool, such as OpenRefine, or Google Sheets or Excel.
• Discussion questions:
  • What were some trends that you noticed in the dataset that you had to address in cleaning?
  • What were some of the challenges you encountered? Where did those challenges originate (dataset, tool, skills, etc.)?

Total recommended time

• 45-60 minutes

Tools/materials required

• Data cleaning software
• Simple, interesting dataset that needs to be cleaned
• Computers for each participant or group
• If virtual, breakout rooms for groups