Selecting columns by name in Pandas

Introduction

We select columns by name to focus on specific data we want to analyze or use. It helps us work only with the information we need.

You want to see only the 'age' and 'salary' columns from a large employee dataset.

You need to create a new table with just the 'name' and 'email' columns for sending newsletters.

You want to analyze sales data but only for the 'region' and 'sales_amount' columns.

You want to remove unnecessary columns before saving a smaller file.

You want to quickly check values in a specific column like 'date' or 'status'.

Syntax

Pandas

df['column_name']
df[['col1', 'col2']]
# df is your DataFrame

Use single brackets [] for one column, double brackets [[]] for multiple columns.

The result for one column is a Series; for multiple columns, it's a DataFrame.

Examples

Selects the column named 'age' as a Series.

Pandas

df['age']

Selects two columns 'name' and 'email' as a new DataFrame.

Pandas

df[['name', 'email']]

Stores selected columns in a new variable and prints the first 5 rows.

Pandas

selected = df[['region', 'sales_amount']]
print(selected.head())

Sample Program

This code creates a small table with people data. It then selects one column 'age' and prints it. Next, it selects two columns 'name' and 'email' and prints that smaller table.

Pandas

import pandas as pd

# Create a sample DataFrame
data = {
    'name': ['Alice', 'Bob', 'Charlie'],
    'age': [25, 30, 35],
    'email': ['[e-mail]', '[e-mail]', '[e-mail]'],
    'salary': [70000, 80000, 90000]
}
df = pd.DataFrame(data)

# Select the 'age' column
age_column = df['age']
print('Age column:')
print(age_column)

# Select multiple columns 'name' and 'email'
contact_info = df[['name', 'email']]
print('\nContact info DataFrame:')
print(contact_info)

Important Notes

If you try to select a column name that does not exist, pandas will give an error.

Using double brackets always returns a DataFrame, which is useful if you want to keep the table format.

Summary

Select columns by their names to work with specific data.

Use single brackets [] for one column (returns Series).

Use double brackets [[]] for multiple columns (returns DataFrame).