Group By in Pandas

Powerful Aggregation

If you're learning pandas, GroupBy is the feature that will make you feel like a real data analyst. It lets you split your data into groups (e.g., by department, city, or category), apply calculations (like sum, mean, count), and combine the results — all in just a few lines of code.

This "split-apply-combine" approach is incredibly useful for real-world tasks like:

• Sales summary by region or product
• Average salary by department
• Customer behavior by age group
• Exam scores by class or gender

Our Sample DataFrame

We'll use a simple company employee dataset:

import pandas as pd

data = {
    'Department': ['Sales', 'Sales', 'Engineering', 'Engineering', 'HR', 'HR', 'Sales', 'Engineering'],
    'Employee': ['Alice', 'Bob', 'Charlie', 'David', 'Eve', 'Frank', 'Grace', 'Heidi'],
    'Salary': [60000, 65000, 90000, 95000, 55000, 58000, 62000, 98000],
    'Experience_Years': [3, 5, 8, 10, 4, 6, 4, 12]
}

df = pd.DataFrame(data)
print(df)

This will display:

	Department	Employee	Salary	Experience_Years
0	Sales	Alice	60000	3
1	Sales	Bob	65000	5
2	Engineering	Charlie	90000	8
3	Engineering	David	95000	10
4	HR	Eve	55000	4
5	HR	Frank	58000	6
6	Sales	Grace	62000	4
7	Engineering	Heidi	98000	12

Basic GroupBy: Average Salary by Department

The simplest way: group by a column and apply an aggregation like mean(), sum(), or count().

# Average salary per department
avg_salary = df.groupby('Department')['Salary'].mean()
print(avg_salary)

Output:

Department	Salary
Engineering	94333.33
HR	56500.00
Sales	62333.33

Multiple Aggregations at Once

Use agg() (or aggregate()) to get several stats together.

# Multiple stats for Salary and Experience
summary = df.groupby('Department').agg({
    'Salary': ['mean', 'sum', 'count'],
    'Experience_Years': ['mean', 'max']
})

print(summary)

Output:

Department	Salary	Experience_Years
mean	sum	count	mean	max
Engineering	94333.33	283000	3	10.00	12
HR	56500.00	113000	2	5.00	6
Sales	62333.33	187000	3	4.00	5

Common Aggregations (Quick Cheat Sheet)

• .mean() — Average
• .sum() — Total
• .count() — Number of rows
• .min() / .max() — Lowest/Highest
• .std() — Standard deviation
• .size() — Count including NaN

Key Tips for Beginners

• GroupBy returns a GroupBy object — always apply an aggregation function afterward.
• Use .reset_index() if you want a regular DataFrame back.
• You can group by multiple columns: df.groupby(['Col1', 'Col2']).
• Perfect for exploratory data analysis (EDA) — run GroupBy early to spot patterns!

Happy Learning !! 🐼