Course contentsShow
← Data Science
Lesson 1172 of 2,14530. Exploratory Data Analysis (EDA) ProcessFree lesson

What is Univariate Analysis?

Understanding analysis of single variables in isolation to reveal distribution, central tendency, and spread.

What is Univariate Analysis?

You'll learn: What univariate analysis is and why examining single variables in isolation is the essential first step in exploratory data analysis.

The Core Idea

Univariate analysis means analyzing one variable at a time, in complete isolation from all others. Think of it like getting to know each player on a team individually before understanding how they work together. You're not looking at relationships or interactions—just understanding what each single column of data looks like on its own.

This is typically your first stop after loading and cleaning data. Before you can understand how variables relate to each other, you need to understand each one independently.

What Univariate Analysis Reveals

When you analyze a single variable, you're trying to answer three fundamental questions:

1. Distribution

What does the overall pattern look like? Is it bell-shaped, skewed, uniform, or something else? Are there multiple peaks?

2. Central Tendency

Where does the "center" of the data sit? What's typical or average?

3. Spread

How much variation exists? Are values tightly clustered or widely scattered? Are there outliers far from the norm?

Why Start Here?

You wouldn't try to understand a conversation in a foreign language without first learning individual words. Similarly, you can't understand multivariate relationships without first grasping what each variable represents on its own. Univariate analysis helps you spot data quality issues, understand variable characteristics, and make informed decisions about which analytical techniques to use next.

Key Takeaway: Univariate analysis examines one variable at a time to understand its distribution, central tendency, and spread—forming the foundation for all subsequent exploratory analysis.