Data & reasoning · Some Python

Find the story in a small dataset

Ask a question, inspect the evidence, and explain what the numbers can—and cannot—tell you.

What you will make

A reproducible notebook with a data comparison, a conclusion, and its limitations.

Before you start

Basic Python and a local Python environment with pandas installed.

Estimated time: 135 minutes. Free reading and open-source tools; no paid dataset is required.

1. Start with an answerable question

Estimated time: 30 minutes.

A good analysis starts with a clear comparison. Decide what one row represents, what you will measure, and which missing values might change your answer.

pandas: selecting data — Learn how to inspect columns and choose rows before drawing conclusions.

Your challenge

Create a small table of invented daily study sessions: date, subject, and minutes. Include at least ten rows and one missing value. Write a question about the data before calculating anything.

Check your understanding

Can you say exactly what one row and each column represent?

A little hint

Try 'Which subject received more recorded study time?' Avoid claims about learning quality when you only measure minutes.

2. Make a comparison you can explain

Estimated time: 60 minutes.

Filtering changes which observations contribute to a result. State the rule and show how much data remains so someone else can reproduce your comparison.

pandas: selecting data — Boolean filters and column selection support a transparent, reproducible comparison.

Your challenge

Select two subjects and compare their recorded minutes. State how you handled the missing value. Include the number of sessions for each subject alongside the totals.

Check your understanding

Could someone reproduce your numbers from the original table and your filtering rule?

A little hint

Filter one subject at a time, inspect the selected rows, and only then calculate the total. A missing value is not automatically zero.

3. Write a conclusion with boundaries

Estimated time: 45 minutes.

A result describes the data you have. It does not automatically explain a cause or generalize to everyone. Limitations are part of a useful answer.

Python: data structures — Keep the original records available so the analysis can be checked against them.

Your challenge

Write a three-sentence finding: the comparison, the supporting numbers, and one limitation. Change one input record, rerun the analysis, and explain whether your conclusion changes.

Check your understanding

Have you avoided claiming that more study minutes caused better learning?

A little hint

Use 'In these invented sessions…' to scope your conclusion. Mention the sample size and the missing observation.

Enable JavaScript to use the private notebook and track self-reported completion. The lessons and source links above are available without JavaScript.