A4.2 Data preprocessing: notes and practice
A4.2 (HL only) covers getting data ready for a model: cleaning missing values, duplicates and outliers, choosing features and reducing the number of dimensions.
Paper 1HL only3 practice questions
What you need to know
Our summary of the topic in our own words, not the IB guide. Points marked HL are for Higher Level only.
- Explain why raw data must be cleaned before training and what happens if it is notHL
- Identify data quality problems such as missing values, duplicates, outliers and inconsistent formatsHL
- Choose a treatment for a given data quality problem and justify itHL
- Explain what feature selection is and why fewer, better features can improve a modelHL
- Identify which features are likely to be informative for a described prediction taskHL
- Explain what dimensionality reduction is and why high-dimensional data is a problemHL
- Describe at a conceptual level how a dimensionality reduction technique projects data into fewer dimensionsHL
Practice questions
Original questions written for the 2027 course, marked as you go. Easiest first.
- Choosing features to predict exam marksWritten · Core
- Cleaning air-quality sensor dataWritten · Stretch
- Cleaning bike-share trip recordsWritten · Stretch
Exam tips for A4.2
- For each data problem, give a treatment and say what could go wrong if you skipped it.
- Explain why fewer, better features can make a model more accurate and faster to train.
- Describe dimensionality reduction at the level of ideas. You are not asked for the maths.
Updated 24 October 2026. Topic codes are used as labels. Everything on this page is written by ExamSpine and is not endorsed by the IB.