Data Analysis with Python
From a messy CSV to a defensible answer, using pandas the way analysts really do.
Tobias Lund
Data scientist · ex-forecasting lead
- Analytics
- SQL
- Pandas
- Python
What you will be able to do
- Load and clean genuinely messy real-world data
- Reshape, group and join without silent row loss
- Spot the analysis mistakes that survive code review
- Visualise a result so the conclusion is unmistakable
- Structure a notebook someone else can rerun
About this course
A hands-on analysis course built entirely on realistic, dirty data. You will clean it, reshape it, join it badly, notice you joined it badly, and fix it. Along the way you will learn the pandas idioms that avoid silent errors, and the habits that make an analysis reproducible by someone who is not you.
Curriculum
3 sections · 9 lessons
- 1Reading files that fight backEncodings, mixed types, dates that are not dates.Preview19 min
- 2Missing data, honestlyWhy it is missing decides what you are allowed to do about it.21 min
- 3Types, categories and memoryMaking a big frame fit and go faster at the same time.16 min
Before you start
- Basic Python — variables, functions, loops
- No statistics background required
Student reviews
4.7 from 3
Daniel Osei
19 Aug 2026Excellent, though it moves quickly
Superb content and no filler at all. I did have to rewatch two lessons at 0.75x. If you are brand new to the subject, budget extra time for the middle section.
11 found this helpful
Ryan Whitfield
19 Aug 2026Worth it for the case study alone
The final project is genuinely realistic — messy inputs, ambiguous requirements, the lot. I have already reused the structure at work twice.
6 found this helpful
Sofía Marchetti
19 Aug 2026Worth it for the case study alone
The final project is genuinely realistic — messy inputs, ambiguous requirements, the lot. I have already reused the structure at work twice.
7 found this helpful