How to Access a Column in a DataFrame with Pandas

In this article we will explore various techniques to access a column in a dataframe with pandas with concise explanations and practical examples.

Method 1: Accessing a Single Column Using Bracket Notation

Bracket notation is the most straightforward method to access a column. Use the syntax df['column_name'] to retrieve the column as a Pandas Series. This method is quick, intuitive, and works for all valid column names.

import pandas as pd

data = {
    'Name': ['John', 'Alice', 'Bob', 'Eve'],
    'Age': [25, 30, 22, 35],
    'Salary': [50000, 55000, 40000, 70000]
}
df = pd.DataFrame(data)

# Accessing the 'salary' column using bracket notation
age_column = df['Salary']
print(age_column)

0    50000
1    55000
2    40000
3    70000
Name: Salary, dtype: int64

Method 2: Accessing a Single Column Using Dot Notation

In addition to bracket notation, you can also access columns using dot notation (df.column_name). This is a more concise and readable approach but can only be used when the column name is a valid Python attribute.

import pandas as pd

data = {
    'Name': ['Michael', 'Sarah', 'David', 'Emma'],
    'Age': [40, 28, 33, 25],
    'Salary': [60000, 62000, 45000, 75000]
}
df = pd.DataFrame(data)

# Accessing the 'Name' column using dot notation
name_column = df.Name
print(name_column)

0    Michael
1      Sarah
2      David
3       Emma
Name: Name, dtype: object

Method 3: Accessing Multiple Columns Using Bracket Notation

You can access multiple columns by passing a list of column names inside the brackets. This returns a new DataFrame containing only the specified columns.

import pandas as pd

data = {
    'Name': ['Michael', 'Sarah', 'David', 'Emma'],
    'Age': [40, 28, 33, 25],
    'Salary': [60000, 62000, 45000, 75000]
}
df = pd.DataFrame(data)

# Accessing 'Name' and 'Salary' columns using bracket notation
subset_columns = df[['Name', 'Salary']]
print(subset_columns)

      Name  Salary
0  Michael   60000
1    Sarah   62000
2    David   45000
3     Emma   75000

Method 4: Accessing Columns by Index Using iloc

If you don't know the column names but know their positions you can use the iloc indexer to access columns by their integer index. This is particularly useful for large datasets or when you need to access columns programmatically.

import pandas as pd

data = {
    'Name': ['Michael', 'Sarah', 'David', 'Emma'],
    'Age': [40, 28, 33, 25],
    'Salary': [60000, 62000, 45000, 75000]
}
df = pd.DataFrame(data)

# Accessing the second column (Age) using iloc
age_column = df.iloc[:, 1]
print(age_column)

0    40
1    28
2    33
3    25
Name: Age, dtype: int64

You can refer this article for detailed explanation: Extracting rows using Pandas .iloc[] in Python

Method 5: Accessing Columns by Condition Using Boolean Indexing

You can access columns based on conditions or filters using boolean indexing. This allows to dynamically select rows that meet specific criteria and access their corresponding columns.

import pandas as pd

data = {
    'Name': ['Michael', 'Sarah', 'David', 'Emma'],
    'Age': [40, 28, 33, 25],
    'Salary': [60000, 62000, 45000, 75000]
}
df = pd.DataFrame(data)

# Accessing rows where Salary is greater than or equal to 60000
high_salary = df[df['Salary'] >= 60000]
print(high_salary)

      Name  Age  Salary
0  Michael   40   60000
1    Sarah   28   62000
3     Emma   25   75000

Method 6: Accessing Columns Using loc for Label-Based Indexing

The loc indexer allows to access rows and columns by their labels. It is more flexible than iloc, as it can be used with both row and column names.

import pandas as pd

data = {
    'Name': ['Michael', 'Sarah', 'David', 'Emma'],
    'Age': [40, 28, 33, 25],
    'Salary': [60000, 62000, 45000, 75000]
}
df = pd.DataFrame(data)

# Accessing rows where 'Age' is greater than 30 using loc
age_above_30 = df.loc[df['Age'] > 30]
print(age_above_30)

      Name  Age  Salary
0  Michael   40   60000
2    David   33   45000

Method 7: Accessing Columns Dynamically

Sometimes, you may need to access columns dynamically based on variables or user input. You can achieve this by using a variable that stores the column name and accessing it using bracket notation.

import pandas as pd

data = {
    'Name': ['Michael', 'Sarah', 'David', 'Emma'],
    'Age': [40, 28, 33, 25],
    'Salary': [60000, 62000, 45000, 75000]
}
df = pd.DataFrame(data)

# Access column name dynamically
column_name = 'Salary'
salary_column = df[column_name]
print(salary_column)

0    60000
1    62000
2    45000
3    75000
Name: Salary, dtype: int64

Hence, For simple column access, bracket or dot notation works best. If you're working with dynamic conditions or large datasets, consider using methods like iloc, loc, or boolean indexing. Experiment with these techniques to find the best approach for your data manipulation tasks.