Before dplyr

The R-package dplyr represents an important milestone in the history of R. Before dplyr existed, data manipulation was not considered to be a strong point of the R system. I even remember very vaguely that even John Chambers was advocating in one of his talks many years ago, that data preparation is better done by some scripting language, like python or perl.

What is it all about

dplyr can be understood as a language of data manipulation. The language consists only of a small number of verbs each designed to perform a well defined task. Each of the verbs is implemented in an R function. Data manipulation processes can be constructed by chaining together sequences of verbs to a pipeline.

Getting started

Introductory slides show the basic usage of dplyr using Andersons Iris data set.

The dplyr package is available through CRAN, hence dplyr can be installed via

install.packages("dplyr")

The introductory vignette to dplyr available through

browseVignettes(package = "dplyr")

demonstrates the application of dplyr to the New York City airport flights dataset.