Glossary
Below is a reference guide for all core terms, libraries, and functions used in R Exploratory Data Analysis.
Tidyverse
A curated collection of R packages designed for data science that share an underlying philosophy, grammar, and data structure. Key packages include ggplot2, dplyr, tidyr, readr, stringr, forcats, lubridate, and purrr.
ggplot2
The standard data visualization package in R, built on the Grammar of Graphics framework. It constructs plots by adding layers (geoms, aesthetics, scales, coordinates, and themes) to a blank data canvas.
Aesthetic Mapping (aes)
An expression that connects a data column in a table to a visual property of a geometric shape on the plot (such as its x-axis position, y-axis position, color, fill, shape, or size).
Geom
Short for geometric object. It defines the visual representation of data points on the plot canvas (e.g. geom_point draws scatter points, geom_line draws lines, geom_bar draws rectangles, geom_boxplot draws box plots).
Stat
Short for statistical transformation. It performs computations or aggregates data points before rendering a geom (e.g., stat_count counts categories for geom_bar, stat_smooth computes regression equations for trend lines).
Facet
A technique to split a single chart canvas into a grid of multiple subplots based on distinct categories of a variable, sharing the same scale for comparison.
coord_flip()
A coordinate system function that swaps the horizontal (x) and vertical (y) axes. It is commonly used to render vertical boxplots horizontally or display long categorical labels legibly.
coord_polar()
A coordinate system function that maps Cartesian coordinates to polar coordinates (angle and distance from the center), wrapping charts radially.
ggsave()
The standard function in ggplot2 to save the active or last rendered plot to a file on disk (in formats like PNG, PDF, JPG, SVG).
dplyr
The standard R package containing a grammar of data transformation verbs. It provides functions to manipulate, filter, summarize, and combine data frames.
Tibble
A modern, enhanced version of R's native data frame structure (data.frame), optimized to print cleanly in the console and avoid silent data conversions.
filter()
A dplyr verb that returns a subset of rows satisfying one or more logical conditions.
select()
A dplyr verb that extracts specific columns from a dataset by name, index, or string helper (like starts_with()).
mutate()
A dplyr verb that creates new columns by applying calculations to existing variables or modifies current columns in-place.
group_by()
A dplyr verb that segments a dataset into groups based on categories in one or more columns. Subsequent summaries or mutations execute within each group.
ungroup()
A dplyr verb that removes grouping constraints from a data frame. It is a critical best practice to add ungroup() to the end of grouping pipelines to avoid unexpected downstream results.
summarize()
A dplyr verb that collapses a table containing many rows into a single row of computed statistics (such as mean, median, sd, n()).
across()
A helper function inside summarize() or mutate() to apply the same function or calculation to multiple selected columns simultaneously.
Window Function
A function that operates on elements relative to their positions, taking vector values as input and returning output values (e.g. ranks, lags, cumulative sums), preserving the row count.
min_rank()
A ranking function that orders values, assigning identical rank numbers to duplicates (ties) and leaving gaps for succeeding positions (e.g. 1st, 1st, 3rd).
dense_rank()
A ranking function that orders values, assigning identical rank numbers to ties without leaving gaps for succeeding positions (e.g. 1st, 1st, 2nd).
lead() and lag()
Window functions that shift a vector's elements offset-up or offset-down. They are commonly used to calculate differences between successive observations (e.g. day-to-day stock changes).
tidyr
A tidyverse package designed to clean and reshape datasets to ensure they conform to tidy data specifications.
pivot_longer()
A tidyr function that reshapes a wide dataset into a long dataset, melting column headers into rows.
pivot_wider()
A tidyr function that reshapes a long dataset into a wide dataset, splitting rows into multiple columns.
separate_wider_delim()
A tidyr function that splits a single text column into multiple distinct columns based on a delimiter character.
Mutating Join
A database operation (such as left_join, inner_join, full_join) that merges columns from a secondary table into a primary table using matching keys.
Filtering Join
A database operation (such as semi_join, anti_join) that filters the rows of a primary table based on whether matching keys exist in a secondary table, without adding columns.
dbplyr
A database driver package that translates standard dplyr pipelines into SQL queries automatically, running calculations on the database server directly (lazy evaluation).
Shiny
An R package that enables developers to build interactive web applications and dashboards directly in R code, linking inputs (widgets) and outputs (plots/tables) reactively.
Reactivity
The programming model behind Shiny apps. It automatically re-runs R code blocks and updates visual outputs on a webpage whenever a user alters input values.