Reference

Last updated on 2026-07-28 | Edit this page

Other Resources and Further Learning

Glossary

argument: A value passed to a function when it is called, used to control what the function does. Some arguments are required; others have default values and are called options.
assignment operator: The <- symbol used in R to assign a value to an object, e.g. weight_kg <- 55. Can be read as “55 goes into weight_kg”. The = sign can also perform assignment but is conventionally reserved for setting function arguments.
class: The type of an R object (e.g. numeric, character, data.frame, factor), which determines how functions handle it. Found using the class() function.
coercion: The automatic conversion of values from one class to another when they are combined, for example when mixed data types are placed in a single vector. Coercion follows a hierarchy: logical → integer → numeric → character.
comment: A remark in a script intended to help human readers understand the code, but ignored when the code runs. Comments in R start with a # character and run to the end of the line.
console: The pane in RStudio where R commands can be typed and are executed immediately, and where results are displayed. Indicated by a > prompt when ready for input, or a + prompt when an incomplete command is still being entered.
coordinates: In the context of a data frame, the row and column position (e.g. surveys[1, 6]) used with square-bracket indexing to extract specific data. Row numbers are specified first, followed by column numbers.
data frame: A tabular data structure in which columns are vectors of equal length, and each column holds a single type of data. The standard structure for tabular data in R, most commonly created by read.csv() or read.table().
data structure: A way of organising and storing data in R. Common data structures include vectors, lists, matrices, data frames, factors, and arrays.
factor: A data structure used to represent categorical data. Factors are stored internally as integers associated with text levels, and can be ordered or unordered.
filtering: Subsetting rows of a data frame based on a logical condition, most commonly with the dplyr function filter(), e.g. filter(surveys, year == 1995).
function: A named, reusable set of instructions that performs a task, usually taking one or more arguments as input and returning a value as output. Running a function is called calling it, e.g. sqrt(10).
geom: Short for “geometric object”; a ggplot2 layer function that determines how data is visually represented in a plot, such as geom_point() for scatter plots, geom_boxplot() for boxplots, or geom_line() for line plots.
aesthetic mapping: In ggplot2, the association between variables in a dataset and visual properties of a plot (position, colour, size, shape), set using the aes() function, e.g. aes(x = weight, y = hindfoot_length).
faceting: A ggplot2 technique that splits a single plot into multiple small panels based on the values of one or more variables, using facet_wrap() or facet_grid().
level: One of a pre-defined set of possible values that a factor can take. By default, R sorts levels alphabetically. Found using the levels() function.
library: A collection of R functions, data, and documentation bundled together as a package; also the name of the function (library()) used to load an installed package into an R session.
missing value: A value that is absent from a dataset, represented in R by NA. Many functions return NA if any input contains missing values unless told to ignore them, typically with the argument na.rm = TRUE.
object: A named “container” in R that stores a value, such as a number, character string, vector, or data frame. Also referred to as a variable. Created using the assignment operator.
option: An argument to a function that has a default value, typically used to alter how the function behaves (e.g. digits in round()). Can be left out to use the default, or specified to override it.
package: A collection of R functions, data, and documentation that extends R’s built-in capabilities. Must be installed once with install.packages() and loaded in each session with library().
pipe: An operator that passes the result of one expression as the first argument to the next, allowing multiple operations to be chained together in a readable sequence. Written as |> in base R (or %>% from the magrittr/tidyverse packages).
pivoting: Reshaping a data frame between “wide” and “long” formats using tidyr functions: pivot_longer() converts columns into rows (wide to long), and pivot_wider() converts rows into columns (long to wide).
reproducibility: The ability for someone else, including your future self, to obtain the same results from the same dataset using the same documented analysis steps, typically achieved by working from saved scripts rather than manual, undocumented actions.
script: A plain text file containing a saved sequence of R commands, which can be run and re-run to reproduce an analysis. Written and edited in the RStudio Source pane.
split-apply-combine: A data analysis strategy in which data is split into groups (using group_by()), a function is applied to each group (e.g. with summarize()), and the results are combined into a single output.
subsetting: Extracting specific elements from a vector or data frame, using numeric indices, logical conditions, or column/row names inside square brackets, or with dplyr functions like select() and filter().
tibble: The tidyverse version of a data frame (class tbl_df), with slightly different default printing and behaviour, typically created by read_csv() rather than read.csv().
value: A piece of information stored and retrieved by an object, such as a number, a character string, or a collection of these, such as a vector.
vector: The most basic data structure in R: an ordered collection of values of the same class, created with the c() function.
working directory: The folder R uses as the default location to look for and save files. Best managed using an RStudio Project, and checked with getwd(). Scripts should not hard-code changes to it with setwd(), since this will not work on someone else’s computer.