DEV Community

Cover image for 📊 Univariate Analysis in Data Science: A Complete Beginner to Pro Guide
Vikas Gulia
Vikas Gulia

Posted on

📊 Univariate Analysis in Data Science: A Complete Beginner to Pro Guide

"Before diving deep into data, start by understanding each variable on its own."

In data science, the first step in understanding a dataset is to analyze one variable at a time. This is called Univariate Analysis.

It is the foundation of Exploratory Data Analysis (EDA) and plays a crucial role in:

  • Spotting data issues
  • Understanding distributions
  • Making modeling decisions

✅ What is Univariate Analysis?

Univariate Analysis is the statistical analysis of a single variable (i.e., “uni” = one).

Goals:

  • Understand the central tendency, spread, and distribution
  • Identify outliers, missing values, and patterns
  • Choose the right preprocessing techniques (e.g., binning, normalization)

🧠 Types of Variables

Univariate analysis depends on the type of variable:

Variable Type Examples Analysis Type
Numerical Age, Salary, Marks Statistical + Visual
Categorical Gender, City, Grade Frequency + Visual

🔢 Univariate Analysis for Numerical Variables

Example: Age of Employees 

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

# Sample data
data = pd.DataFrame({
    "Age": [22, 25, 24, 29, 30, 23, 22, 45, 32, 41, 38, 27]
})

# Summary Statistics
print(data["Age"].describe())
Enter fullscreen mode Exit fullscreen mode

Output: 

count    12.000000
mean     30.250000
std       7.909809
min      22.000000
25%      23.250000
50%      27.500000
75%      32.750000
max      45.000000
Enter fullscreen mode Exit fullscreen mode

Visualizations

Histogram 

sns.histplot(data["Age"], bins=6, kde=True)
plt.title("Age Distribution")
plt.xlabel("Age")
plt.show()
Enter fullscreen mode Exit fullscreen mode

Box Plot 

sns.boxplot(x=data["Age"])
plt.title("Boxplot of Age")
plt.show()
Enter fullscreen mode Exit fullscreen mode

📌 Boxplots help detect outliers.
📌 Histograms help understand the shape of distribution (normal, skewed, etc.)

🟦 Univariate Analysis for Categorical Variables

Example: Department 

data = pd.DataFrame({
    "Department": ["HR", "IT", "IT", "Sales", "HR", "IT", "Sales", "Sales", "IT"]
})

# Frequency Table
print(data["Department"].value_counts())

# Bar Plot
sns.countplot(x="Department", data=data)
plt.title("Department Distribution")
plt.show()
Enter fullscreen mode Exit fullscreen mode

Output: 

IT       4
Sales    3
HR       2
Enter fullscreen mode Exit fullscreen mode

📌 Bar charts are great for visualizing categorical variable frequencies.

📊 Summary Table of Techniques

Variable Type Technique Visualization
Numerical mean, median, std histogram, boxplot
Categorical value_counts(), mode bar plot, pie chart

🧪 When and Why to Use Univariate Analysis?

Use Case Why Important
Data Cleaning Detect missing values and outliers
Feature Engineering Understand variable behavior
Model Selection Identify skewed or non-normal distributions
Business Insights Understand customer age, sales region, etc.

🚫 Common Mistakes

  • Ignoring skewness and directly applying normal assumptions
  • Not visualizing the data before modeling
  • Not treating outliers (can mislead models)

📁 Real-Life Examples

  • 🎯 E-commerce: Analyze purchase amount distribution
  • 🏥 Healthcare: Examine age distribution of patients
  • 🏢 HR Analytics: Check gender or department distribution
  • 📈 Finance: Analyze transaction amount or loan default categories

🧠 Final Thoughts

Univariate analysis is the first diagnostic tool you should apply to any dataset. It’s simple, yet incredibly powerful. It helps data scientists make informed decisions and avoid costly mistakes in preprocessing and modeling.

“If you don’t understand your variables, your model won’t either.”

Top comments (0)