Data Science: final assessment

This covers the whole course. Take it after the last module.

Final quiz

1. What is the first step of a data science project?
2. Which value most likely means missing in an age column?
3. Which summary is least affected by outliers?
4. A test set is used to…
5. Accuracy is (TP + TN) divided by…
6. Tidy data has…
7. An inner join keeps…
8. Percentage change from 80 to 100 is…
9. Three numeric columns give how many correlation pairs?
10. A classifier flags 10 rows, 8 correctly, out of 16 real positives. Its precision and recall are…
11. Why print empty bins in a histogram?
12. Ice cream sales and sunburns are correlated. The best explanation is…
13. A model's test error is far worse than its training error. The likely problem is…
14. Which split rule is reproducible?
15. Monthly sales jump every December. To judge growth, compare December with…
16. Half the rows were dropped for missing income. What must the report say?

Practical assessment: A survey data pipeline

Without using hints, write a plain-Python pipeline that cleans a messy survey export, summarises it by group, checks one relationship, and prints a short report with a cleaning log.

Final project

A sales data analysis: A shop exports a year of messy sales records. Clean them, answer which products, regions and months drive sales, and write a short report with the cleaning log.

How it is evaluated

To complete this course