Exploratory Analysis for Data Sciences
Exploratory data analysis, or EDA, is the first look at a dataset using summary statistics and charts to understand its shape, quality and relationships before modelling. It also includes data quality checks and documenting what each field means, so findings and assumptions can be reviewed by others.
Stage: Data Wrangling and Exploration · Part of the Data Science roadmap · Last reviewed 19 November 2026
What to learn in Exploratory Analysis
- Summarise a dataset with describe, value counts and missingness tables
- Choose a histogram, box plot or scatter plot for the question being asked
- Spot Simpson's paradox by comparing aggregate and segment results
- Check for duplicates, impossible values and inconsistent categories
- Write a short data dictionary and list of assumptions
Exploratory Analysis interview questions
What would you check first when given an unfamiliar dataset?
Check its size, column types, the time range, missing values, duplicate rows and what one row represents. Then look at distributions and the target variable. Confirming the grain and source early prevents analysing the wrong thing.
What is Simpson's paradox?
It is when a trend in the combined data reverses or disappears once the data is split into groups. It happens when group sizes differ and a confounding variable drives both groups, so always check key segments before trusting an overall number.
Where Exploratory Analysis fits in the Data Science roadmap
Exploratory Analysis belongs to the Data Wrangling and Exploration stage (intermediate level), which takes about 5 hours. Know your data.
Other units in this stage
Back to the full Data Science roadmap