How to compute proportions in R

statistics

proportion

Learn compute proportions in r with clear examples and explanations.

Published

March 26, 2026

Introduction

The count() function in dplyr is one of the most useful tools for quickly summarizing categorical data in R. It provides an efficient way to count observations by groups and can be combined with other functions to calculate proportions and percentages. This tutorial shows how to use count() effectively with the Palmer Penguins dataset.

Setup

First, let’s load the required packages and prepare our data:

library(tidyverse)
library(palmerpenguins)

We’ll remove any rows with missing values to ensure clean counts:

penguins_clean <- penguins |>
  drop_na()

This gives us a complete dataset with 333 penguin observations to work with.

Basic Counting

The simplest use of count() is to count observations in a single variable:

penguins_clean |>
  count(species)

This returns a tibble showing how many penguins of each species are in our dataset. The count() function automatically creates a column called n with the counts.

Adding Proportions

To convert counts to proportions, we can add a mutate() step:

penguins_clean |>
  count(species) |>
  mutate(prop = n / sum(n))

This calculates what fraction of all penguins each species represents. The proportions will sum to 1.0.

Counting Multiple Variables

You can count combinations of multiple variables by listing them in count():

penguins_clean |>
  count(species, sex)

This shows the count for each combination of species and sex, giving us six rows (3 species × 2 sexes).

Grouped Proportions

To calculate proportions within each species (rather than across all penguins), use group_by():

penguins_clean |>
  count(species, sex) |>
  group_by(species) |>
  mutate(proportion = n / sum(n))

Now the proportions represent the male/female split within each species, and each species’ proportions sum to 1.0.

Alternative Approach with .by

Modern dplyr offers a cleaner syntax using the .by argument:

penguins_clean |>
  count(species, sex) |>
  mutate(proportion = n / sum(n), .by = species)

This produces the same result as the previous example but avoids explicit grouping and ungrouping.

Overall Proportions vs Grouped Proportions

It’s important to understand the difference between overall and grouped proportions:

# Overall proportions (across all penguins)
penguins_clean |>
  count(species, sex) |>
  mutate(overall_prop = n / sum(n))

This calculates what fraction each species-sex combination represents of the entire dataset. These proportions will be much smaller since they’re divided by the total number of penguins (333) rather than by species totals.

Summary

The count() function is essential for exploratory data analysis in R. Key takeaways: - Use count() alone for simple frequency tables - Combine with mutate() to add proportions - Count multiple variables to examine combinations - Use .by or group_by() to calculate proportions within groups - Always consider whether you want overall or grouped proportions for your analysis

--- title: "How to compute proportions in R" description: "Learn compute proportions in r with clear examples and explanations." date: 2026-03-26 categories: ['statistics', 'proportion'] format: html: code-fold: false code-tools: true --- ## Introduction The `count()` function in dplyr is one of the most useful tools for quickly summarizing categorical data in R. It provides an efficient way to count observations by groups and can be combined with other functions to calculate proportions and percentages. This tutorial shows how to use `count()` effectively with the Palmer Penguins dataset. ## Setup First, let's load the required packages and prepare our data: ```r library(tidyverse) library(palmerpenguins) ``` We'll remove any rows with missing values to ensure clean counts: ```r penguins_clean <- penguins |> drop_na() ``` This gives us a complete dataset with 333 penguin observations to work with. ## Basic Counting The simplest use of `count()` is to count observations in a single variable: ```r penguins_clean |> count(species) ``` This returns a tibble showing how many penguins of each species are in our dataset. The `count()` function automatically creates a column called `n` with the counts. ## Adding Proportions To convert counts to proportions, we can add a `mutate()` step: ```r penguins_clean |> count(species) |> mutate(prop = n / sum(n)) ``` This calculates what fraction of all penguins each species represents. The proportions will sum to 1.0. ## Counting Multiple Variables You can count combinations of multiple variables by listing them in `count()`: ```r penguins_clean |> count(species, sex) ``` This shows the count for each combination of species and sex, giving us six rows (3 species × 2 sexes). ## Grouped Proportions To calculate proportions within each species (rather than across all penguins), use `group_by()`: ```r penguins_clean |> count(species, sex) |> group_by(species) |> mutate(proportion = n / sum(n)) ``` Now the proportions represent the male/female split within each species, and each species' proportions sum to 1.0. ## Alternative Approach with .by Modern dplyr offers a cleaner syntax using the `.by` argument: ```r penguins_clean |> count(species, sex) |> mutate(proportion = n / sum(n), .by = species) ``` This produces the same result as the previous example but avoids explicit grouping and ungrouping. ## Overall Proportions vs Grouped Proportions It's important to understand the difference between overall and grouped proportions: ```r # Overall proportions (across all penguins) penguins_clean |> count(species, sex) |> mutate(overall_prop = n / sum(n)) ``` This calculates what fraction each species-sex combination represents of the entire dataset. These proportions will be much smaller since they're divided by the total number of penguins (333) rather than by species totals. ## Summary The `count()` function is essential for exploratory data analysis in R. Key takeaways: - Use `count()` alone for simple frequency tables - Combine with `mutate()` to add proportions - Count multiple variables to examine combinations - Use `.by` or `group_by()` to calculate proportions within groups - Always consider whether you want overall or grouped proportions for your analysis