Analysts often spend 50-80% of their time preparing and transforming data sets before they begin more formal analysis work. This video tutorial shows you how to streamline your code—and your thinking—by introducing a set of principles and R packages that make this work much faster and easier. Garrett Grolemund, Data Scientist and Master Instructor at RStudio, demonstrates how R and its packages help you tackle three main issues:
- Data Manipulation. Data sets contain more information than they display. By transforming your data, you can reveal a wealth of descriptive statistics, group level observations, and hidden variables. R’s dplyr package provides optimized functions to help you transform data, as well as a pipe syntax that makes R code more concise and intuitive.
- Data Tidying. Data sets come in many formats, but R prefers just one. R runs quickly and intuitively when your data is stored in the tidy format, a layout that allows vectorized programming. R’s tidyr package reshapes the layout of your data sets, making them tidy while preserving the relationships they contain.
- Data Visualization. The structure of data visualizations parallels the structure of data sets. Once your data is tidy, visualizations become straightforward: each observation in your dataset becomes a mark on a graph, each variable becomes a visual property of the marks. The result is a grammar of graphics that lets you create thousands of graphs. R’s ggvis package implements the grammar, providing a system of data visualization for R.
Garrett Grolemund is a Data Scientist and Master Instructor at RStudio. Garrett maintains the lubridate R package and is the author of Hands-On Programming with R and the upcoming Data Science with R (both O’Reilly books).