dplyr

dplyr is an R package for data manipulation that provides a cohesive set of verbs for common data-wrangling tasks. It is part of the tidyverse, a collection of packages designed for consistent, human-readable data analysis. dplyr operates primarily on data frames and tibbles, emphasizing readable, intent-driven code in data workflows.

The core verbs include select, filter, arrange, mutate, and summarize (summarise). Grouping with group_by enables operations by subset, and join functions such as left_join, inner_join, right_join, full_join, anti_join, and semi_join combine data from multiple tables. Additional helpers include distinct, rename, and slice. The across function provides a way to apply a function to multiple columns in a single call.

Code is typically chained with the pipe operator %>% to pass the result of one operation to the next, improving readability and mimicking a sequence of data transformations. dplyr supports both in-memory data frames and tibbles, and can work with database backends via dbplyr, which translates dplyr verbs into SQL queries that run remotely.

Performance in dplyr is optimized for typical data analysis tasks on moderately sized data. While it can handle large data sets, specialized tools such as data.table may offer different performance characteristics. The package emphasizes a tidy data approach and consistency across data sources.

Developed by Hadley Wickham and contributors, dplyr is a core component of the tidyverse. It is open-source and cross-platform, with extensive documentation and community examples that describe both basic usage and advanced techniques.