r-sprache

R

In short: A programming language specifically for statistics and data analysis — widely used in science, research and bioinformatics.

In more detail: R was developed by statisticians for statisticians and thereby offers extremely extensive built-in functions for data analysis, visualisation and statistical models, often more directly than in Python. In practice, R mainly competes with Python + Pandas/NumPy for data science tasks.

In Depth

A statistics-native language design

data <- read.csv("survey.csv")
mean_value <- mean(data$age, na.rm = TRUE)
model <- lm(income ~ age + years_of_education, data = data)
summary(model)
plot(data$age, data$income)

R was developed in the 1990s (as a free reimplementation of the older, commercial language S) specifically for statistical calculations, not as a general-purpose programming language — accordingly, many core operations (vectors, statistical tests, model fitting like lm() for linear regression in the example above) feel more direct and compact than in generic languages like Python, where the same functionality is retrofitted via additional libraries like NumPy/SciPy, instead of being a fixed part of the language itself.

CRAN as the central package archive

The huge CRAN package archive (“Comprehensive R Archive Network”) covers practically every statistical method ever published in academic research — statisticians frequently publish an R package directly alongside a new scientific method, which is why R is often the fastest way to get brand-new statistical methods not yet ported to other languages.

Division of labour between R and Python

In practice, a rough, not always sharply defined division of labour has become established: R still dominates in academic statistics, biostatistics/epidemiology (e.g. clinical trials), and wherever very specialised, cutting-edge statistical methods are needed. Python (with Pandas/NumPy), by contrast, has the upper hand in industrial software development and deploying machine learning models into production systems — not least because Python, as a general-purpose language, is easier to integrate into larger software systems (web backends, APIs), while R is primarily designed for the actual analysis, not for production use in a larger application.

RStudio as the standard development environment

By far the most popular development environment for R is RStudio (now part of Posit), which combines a code editor, an interactive console, a variable overview and plot display in a single interface — for many statisticians and researchers, practically inseparable from R itself, similar to how certain IDEs hold a quasi-standard position in other language ecosystems.

See also: Data Science, Pandas, NumPy