What Are Pivot Tables, and How Do You Create Them in Pandas?

A Guide to Pivot Tables for Beginner Python Data Analysts

Photo by Andrzej Kryszpiniuk on Unsplash

Pivot Table Explanation

Pivot tables go hand-in-hand with data analytics, but understanding them can be challenging. In this article, I’ll explain them methodically, which may clarify the concept if you have found them difficult in the past.

In a pivot table, each row represents a group of aggregated data, and each column represents a subset of that aggregated data for a particular attribute value.

You can think of creating a pivot table in two stages:

1. First, create the row-wise aggregation of data based on a column (this is the same thing as simply grouping data by a particular column).
2. Second, spread the grouped data over multiple columns by using a separate column to identify the subsets of each group.

Now let’s walk through each stage with visual examples. It will make more sense then!

The Sample Data

Here’s our sample data, a table that tracks video game scores by player, game, and arcade.