1 · Explore a real dataset
This lesson walks the complete first-contact workflow every data scientist runs on a new dataset: load → inspect → summarise → clean → group → visualise → question.
Work through the notebook below cell by cell. Don’t just run it — read the outputs. The skill being trained is noticing things.
The DataLab's lessons were updated since your last visit, so this notebook now shows the new version. Your saved portfolio is not affected.
The notebook above runs entirely in your browser — nothing is installed and nothing leaves your device. Your changes are kept in this browser. First load takes a few seconds while Python starts.
Checkpoint
Section titled “Checkpoint”You should now be able to answer, with code:
- How many rows and columns does a dataset have?
- Which columns have missing values, and what did you do about them?
- What’s the average of a column per group (e.g. per country)?
- Is a correlation positive, negative, or meaningless — and why does correlation ≠ causation?
- In the ethics lab, only one step of the country ranking was bigger than the margin of error. What headline can this data actually support?
- Why is a group of about a dozen people a problem both for accuracy and for privacy?
Prove it: the graded exercise asks you six questions about this exact dataset — with an instant checker.
Co-funded by the European Union under the Digital Europe Programme — projectEURIDICE, Grant Agreement No. 101123121. Views and opinions expressed are those of the author(s) only and do not necessarily reflect those of the European Union or the European Commission. Neither the European Union nor the granting authority can be held responsible for them.
Built by IIIM — Icelandic Institute for Intelligent Machines. Learning materials are licensedCC BY 4.0.