Introduction to Packages

An R package is a bundled unit of reusable code and resources that can be installed and loaded into an R session.

A package typically contains:

• R functions
• Preloaded datasets
• Help documentation
• Compiled C/C++/Fortran code (optional)
• Tests and examples

Packages allow:

• Code reuse
• Faster development
• Access to advanced algorithms
• Standardized and tested solutions
• Community-driven improvements

Examples of tasks done using packages:

• Data manipulation → dplyr
• Data visualization → ggplot2
• Machine learning → caret
• Web apps → shiny
• Statistical modeling → lme4

Base R comes with several default packages that are automatically available.

Examples:

• base
• stats
• graphics
• utils
• methods

Check loaded base packages:

search()

CRAN is the official repository for R packages.

Features:

• Thousands of packages
• Peer-reviewed submissions
• Version control
• Platform support (Windows, macOS, Linux)

Website: https://cran.r-project.org

Installing from CRAN

Use install.packages().

install.packages("ggplot2")

This installs the package on your system.

Installing Multiple Packages

install.packages(c("dplyr", "tidyr", "readr"))

Choosing a CRAN Mirror

chooseCRANmirror()

Using library()

Loads the package into the current session.

library(ggplot2)

Using require()

Returns TRUE/FALSE if package is loaded.

require(dplyr)

You install once, but load every session.

List All Installed Packages

installed.packages()

Check if a Package is Installed

"ggplot2" %in% rownames(installed.packages())

You can access functions using ::.

ggplot2::ggplot(mtcars, ggplot2::aes(wt, mpg))

Useful when:

• Avoiding name conflicts
• Using a single function only

Help for a Package

help(package = "dplyr")

Help for a Function

?filter

Browse Package Vignettes

browseVignettes("dplyr")

Keep packages up to date.

update.packages()

Update specific package:

install.packages("ggplot2")

Uninstall packages you no longer need.

remove.packages("ggplot2")

dplyr

Data manipulation:

• filter()
• select()
• mutate()
• summarise()

ggplot2

Data visualization using grammar of graphics.

ggplot(mtcars, aes(wt, mpg)) + geom_point()

tidyr

Data reshaping:

• pivot_longer()
• pivot_wider()

shiny

Build interactive web applications.

readr

Fast data import/export.

The tidyverse is a collection of related packages designed for data science.

Includes:

• ggplot2
• dplyr
• tidyr
• readr
• purrr
• stringr

Install tidyverse:

install.packages("tidyverse")

Load tidyverse:

library(tidyverse)

Sometimes two packages have functions with the same name.

Example:

• filter() in stats
• filter() in dplyr

Solution:

dplyr::filter()

Advanced users can create their own packages to:

• Share reusable code
• Distribute tools
• Organize large projects

Tools:

• devtools
• roxygen2
• usethis

install.packages("dplyr")
library(dplyr)

data <- data.frame(
  name = c("Alice", "Bob"),
  score = c(85, 90)
)

filter(data, score > 85)

• Forgetting to load package after installing
• Name conflicts between packages
• Installing packages repeatedly
• Using outdated package versions

Packages are the backbone of R’s ecosystem. They allow users to extend R’s capabilities, reuse high-quality code, and perform complex tasks easily. Understanding how to install, load, manage, and use packages is essential for effective R programming and data science work.