Altcademy - a Forbes magazine logo Best Coding Bootcamp 2023

How to drop null values in Pandas

Understanding Null Values

Before we dive into the specifics of how to drop null values using Pandas, it's important to understand what null values are and why they might appear in your data. In the realm of programming and data analysis, a null value represents missing or undefined data. It's like having an empty cell in a spreadsheet. In Python's Pandas library, null values can be represented as None or NaN (Not a Number).

Why Remove Null Values?

Null values can be problematic because they can skew your analysis, give you incorrect results, or even cause errors in your code. Think of it like trying to calculate the average grade in a class but some of the grades haven't been entered yet. If you don't account for these missing grades, the average you calculate won't be accurate.

Getting Started with Pandas

To begin working with Pandas, you need to import the library. If you haven't installed Pandas yet, you can do so using pip:

pip install pandas

Once installed, you can import it into your Python script like so:

import pandas as pd

Creating a DataFrame with Null Values

Let's start by creating a simple DataFrame with some null values. A DataFrame is one of the primary data structures in Pandas, similar to a table in a database or a spreadsheet.

import pandas as pd
import numpy as np  # Numpy is often used alongside Pandas for numerical operations

# Create a simple DataFrame
data = {
    'Name': ['Anna', 'Bob', 'Catherine', 'David', 'Emma'],
    'Age': [28, None, 34, 29, None],
    'Salary': [70000, 48000, np.nan, 54000, 65000]

df = pd.DataFrame(data)

Identifying Null Values

Before you remove null values, you often want to know where they are. Pandas provides the isnull() method, which returns a DataFrame where each cell is either True (if the original cell was null) or False.

# Check for null values
null_mask = df.isnull()

Removing Null Values

Pandas provides several methods for dealing with null values. The two most common are dropna() and fillna(). In this blog, we'll focus on dropna() which is used to drop rows or columns that contain null values.

Dropping Rows with Null Values

The simplest use of dropna() will drop any row that contains at least one null value.

# Drop rows with any null values
df_no_null_rows = df.dropna()

Dropping Columns with Null Values

Alternatively, you can drop columns that contain null values by specifying the axis parameter.

# Drop columns with any null values
df_no_null_columns = df.dropna(axis=1)

Dropping Rows with All Null Values

If you only want to drop rows where all values are null, you can use the how parameter.

# Drop rows where all values are null
df_no_all_null_rows = df.dropna(how='all')

Dropping Rows with Null Values in Specific Columns

Sometimes, you may want to drop rows based on null values in specific columns. Use the subset parameter for this.

# Drop rows where 'Age' is null
df_no_null_age = df.dropna(subset=['Age'])


You can also use the thresh parameter to specify a minimum number of non-null values for the row/column to be kept.

# Keep only the rows with at least 2 non-NA values
df_thresh = df.dropna(thresh=2)

Inplace Deletion

By default, dropna() returns a new DataFrame. If you want to modify your original DataFrame in place, use the inplace parameter.

# Drop rows with any null values in the original DataFrame

Handling Null Values Intuitively

Imagine you have a basket of fruits with some rotten ones. You have several choices: remove the rotten fruits (dropna), replace them with fresh ones (fillna), or decide if a fruit is too rotten based on how many spots it has (thresh). In data analysis, "rotten fruits" are your null values, and you have similar choices for dealing with them.


Dealing with null values is like gardening; you need to identify the weeds before you can remove them. By using Pandas' dropna() method, you can keep your garden of data clean and ready for analysis. Remember, dropping null values is not always the best approach - sometimes, replacing them with meaningful data can be more appropriate. However, when the situation calls for it, now you have the tools to drop null values effectively, ensuring that your data analysis is based on the most complete and accurate information available. Keep practicing, and soon dealing with null values will become as natural as watering plants in your garden.