How to use nest() to create nested dataframes in R

tidyr

tidyr nest()

Learn use nest() to create nested dataframes in r with clear examples and explanations.

Published

March 26, 2026

Introduction

The nest() function from tidyr is a powerful tool for creating nested data structures in R. It allows you to organize your data by grouping rows into list-columns, making it easier to work with complex datasets and perform operations on subsets of your data. This technique is particularly useful when you want to apply models or functions to different groups within your dataset.

Setup

Let’s start by loading the necessary packages and exploring our dataset:

library(tidyverse)
library(palmerpenguins)

The palmerpenguins dataset contains information about three penguin species. Let’s take a quick look at the structure:

penguins |> head()

This gives us a preview of the data, showing variables like species, island, bill measurements, and other characteristics.

Basic Data Grouping

Before diving into nesting, let’s see how many penguins we have for each species:

penguins |> 
  group_by(species) |>
  summarize(n = n())

This shows us the count for each species: Adelie, Chinstrap, and Gentoo penguins.

Simple Nesting by Species

Now let’s create our first nested structure by grouping penguins by species:

penguins |>
  group_by(species) |>
  nest()

This creates a tibble with one row per species and a list-column called data containing all the other variables for each species. Each element in the data column is itself a tibble with the observations for that species.

Nesting with Multiple Groups

We can nest by multiple variables to create more granular groupings:

penguins |>
  group_by(species, island) |>
  nest()

This creates separate nested datasets for each combination of species and island, giving us more specific subsets to work with.

Selective Nesting with Column Specification

Instead of nesting all remaining columns, we can specify exactly which columns to include in the nested data:

penguins |> 
  nest(data = c(bill_length_mm:year))

This approach nests only the columns from bill_length_mm through year, keeping species and island as regular columns in the main tibble.

Nesting Most Columns

We can also nest a large range of columns while keeping key identifiers separate:

penguins |> 
  nest(data = c(island:year))

This nests everything from island to year, leaving only species as the grouping variable in the main tibble.

Summary

The nest() function transforms your data into a nested structure where each row represents a group and contains a list-column of tibbles. This approach is invaluable when you need to apply functions or models to different subsets of your data, perform group-wise operations, or organize complex datasets in a more manageable way. The nested format works seamlessly with purrr functions like map() for iterative operations across groups.

--- title: "How to use nest() to create nested dataframes in R" description: "Learn use nest() to create nested dataframes in r with clear examples and explanations." date: 2026-03-26 categories: ['tidyr', 'tidyr nest()'] format: html: code-fold: false code-tools: true --- ## Introduction The `nest()` function from tidyr is a powerful tool for creating nested data structures in R. It allows you to organize your data by grouping rows into list-columns, making it easier to work with complex datasets and perform operations on subsets of your data. This technique is particularly useful when you want to apply models or functions to different groups within your dataset. ## Setup Let's start by loading the necessary packages and exploring our dataset: ```r library(tidyverse) library(palmerpenguins) ``` The palmerpenguins dataset contains information about three penguin species. Let's take a quick look at the structure: ```r penguins |> head() ``` This gives us a preview of the data, showing variables like species, island, bill measurements, and other characteristics. ## Basic Data Grouping Before diving into nesting, let's see how many penguins we have for each species: ```r penguins |> group_by(species) |> summarize(n = n()) ``` This shows us the count for each species: Adelie, Chinstrap, and Gentoo penguins. ## Simple Nesting by Species Now let's create our first nested structure by grouping penguins by species: ```r penguins |> group_by(species) |> nest() ``` This creates a tibble with one row per species and a list-column called `data` containing all the other variables for each species. Each element in the `data` column is itself a tibble with the observations for that species. ## Nesting with Multiple Groups We can nest by multiple variables to create more granular groupings: ```r penguins |> group_by(species, island) |> nest() ``` This creates separate nested datasets for each combination of species and island, giving us more specific subsets to work with. ## Selective Nesting with Column Specification Instead of nesting all remaining columns, we can specify exactly which columns to include in the nested data: ```r penguins |> nest(data = c(bill_length_mm:year)) ``` This approach nests only the columns from `bill_length_mm` through `year`, keeping `species` and `island` as regular columns in the main tibble. ## Nesting Most Columns We can also nest a large range of columns while keeping key identifiers separate: ```r penguins |> nest(data = c(island:year)) ``` This nests everything from `island` to `year`, leaving only `species` as the grouping variable in the main tibble. ## Summary The `nest()` function transforms your data into a nested structure where each row represents a group and contains a list-column of tibbles. This approach is invaluable when you need to apply functions or models to different subsets of your data, perform group-wise operations, or organize complex datasets in a more manageable way. The nested format works seamlessly with purrr functions like `map()` for iterative operations across groups.