R Cheat Sheet

Learn R from the basics →

Basics

x <- 42: Assign a value
class(x) / typeof(x): Inspect a value's type
NA / NULL / TRUE / FALSE: Missing value, nothing, logicals
?mean / help(mean): Open the help page
install.packages("dplyr"); library(dplyr): Install and load a package

Vectors

v <- c(3, 1, 2): Create a vector
1:10 / seq(0, 1, by = 0.25): Sequences
v[1] / v[-1] / v[v > 1]: First element / all but first / filter (1-based)
length(v) / sum(v) / mean(v): Length and summaries
v * 2: Vectorized arithmetic
sort(v) / rev(v) / unique(v): Reorder and deduplicate

Strings

paste("a", "b") / paste0("a", "b"): Join with / without a space
sprintf("%.2f", 3.14159): Format numbers
nchar(s) / toupper(s) / substr(s, 1, 3): Length, case, substring
grepl("ab", s) / gsub("a", "o", s): Match / replace with regex

Control Flow & Functions

if (x > 0) { } else { }: Conditional (single value)
ifelse(v > 0, "pos", "neg"): Vectorized conditional
for (i in 1:5) { }: For loop
area <- function(w, h = 1) w * h: Function with a default argument
\(x) x^2: Anonymous function (R 4.1+)
x |> sqrt() |> round(2): Native pipe (R 4.1+)

Lists & Data Frames

l <- list(name = "Ada", age = 36); l$name: Named list
df <- data.frame(x = 1:3, y = c("a", "b", "c")): Create a data frame
head(df) / str(df) / summary(df): Inspect a data frame
df$x / df[df$x > 1, ]: A column / filter rows
read.csv("file.csv"): Read a CSV file

dplyr

filter(df, age > 30): Keep rows
select(df, name, age): Keep columns
mutate(df, age2 = age * 2): Add or change columns
arrange(df, desc(age)): Sort rows
df |> group_by(city) |> summarise(n = n(), avg = mean(age)): Grouped summaries

ggplot2

ggplot(df, aes(x = height, y = weight)) + geom_point(): Scatter plot
geom_line() / geom_col() / geom_histogram(): Line, bar, histogram
aes(colour = group): Map a variable to colour
labs(title = "...", x = "...", y = "..."): Titles and axis labels
ggsave("plot.png", width = 6, height = 4): Save the last plot

Statistics

mean(x, na.rm = TRUE) / median(x) / sd(x): Centre and spread
table(x) / quantile(x): Counts / quantiles
cor(x, y): Correlation
model <- lm(y ~ x, data = df); summary(model): Linear regression
t.test(x, y): Two-sample t-test