Overview — What are R, RStudio, and jstats?

In short, R and RStudio are the free tools you run everything inside, and jstats is an add-on package — the one these guides are about. The two short sections below say what each of these is; if you already work in R, skim them and move on.

R and RStudio

R is the engine — the free software that does the actual computing and statistics. RStudio is the dashboard you drive it from: a friendlier window onto R, with a place to write your commands, see your results, view your data, and keep your work organized. You install R first, then RStudio on top of it — and from then on you open RStudio, not R. Most people never open R directly at all.

jstats and other packages

R already does a great deal on its own — R as it comes, before you add anything, is usually called base R. But much of R’s power comes from packages: add-ons that teach R new tricks, each a bundle of extra commands that someone has written and shared. Thousands exist, and they tend to fall into a few broad types:

  1. Some take results computed by base R and lay them out in a cleaner, more organized form — often with far less code.
  2. Some import and export data in other programs’ formats — haven, for instance, reads and writes SPSS, Stata, and SAS files.
  3. Some turn base R’s results into graphs and charts — ggplot2 is the best known.
  4. Some add whole analyses base R doesn’t include on its own — for example, mice for filling in missing data, or packages for mixed models and survival analysis.

jstats is one of these add-ons — but rather than specializing in a single one of these jobs, it draws the most useful pieces from across the whole range and bundles them into one more comprehensive package, aimed at the things social scientists do most often. The practical payoff: instead of installing and loading several separate packages just to get going, you can lean on jstats for most of your early work and add other packages as your needs grow. Because jstats stays close to base R rather than inventing its own dialect, those other packages slot in alongside it naturally when the time comes.

Here’s how jstats draws on each type. Most of what it does is type 1: it takes the analyses social scientists use most — descriptives, frequencies, t-tests, ANOVA, regression, correlation, and more — and presents the results in a clean, familiar layout like the one you’d see in commercial statistical software, all from a single short command. It also does type 2 directly, reading and writing labelled SPSS, Stata, and SAS files so they work without a separate import step; a bit of type 3, making basic graphs more quickly and simply than reaching for ggplot2 when you don’t need its full range of options; and the least of type 4 so far, though it adds a few things base R doesn’t include — for example, Gelman (2008) standardized regression coefficients. Throughout, it leans on base R’s own computational engines underneath, so what you learn stays close to base R.

Next steps

One reassurance before you start, because it matters more than it sounds: nothing you type into R can harm your computer. The worst outcome of any command is a red error message saying R didn’t understand — and errors aren’t a sign you’re doing badly. They’re a normal part of how everyone works in R, beginners and forty-year veterans alike. The setup steps ahead — installing two programs and a package — are honestly the clumsiest part of the whole journey, and things get easier the moment they’re done. Work through them patiently, lean on the Troubleshooting page if anything snags, and once you’re in, experiment freely.

With this background in mind, you’re ready for Step 2, where we install R and RStudio, and then Step 3, a brief tour of RStudio. Already comfortable in RStudio? Skip ahead to Step 4 (installing jstats) and Step 5 (the Quick Start).

That’s all the background you need to get going. If you’d like a little more first — where the names came from, R’s history, or a closer look at some of R’s best-known packages — the optional boxes below are there to open or skip as you please.

R’s roots reach back to S, a language created at Bell Labs in the 1970s for statistics and data analysis. (Bell Labs had a taste for one-letter names — it’s the same lab behind the C programming language and the Unix operating system — and S nods to statistics.) S was later turned into commercial software called S-PLUS. Then, in the early 1990s, two statisticians at the University of Auckland in New Zealand — Ross Ihaka and Robert Gentleman — wrote a free, open-source language of their own in the spirit of S and named it R: the letter just before S, and the shared first initial of both their names. R has been developed ever since by a worldwide volunteer group, the R Core Team, and stays free for everyone — which is a big part of why it has grown so many add-on packages.

jstats quietly continues that naming tradition. It pairs the j of its author’s first name with stats, and every command the package adds begins with that same j — a small signal that lets you tell at a glance which functions come from jstats rather than from base R or another package.

The best-known friendlier layer over R is the tidyverse — a family of packages (dplyr for data wrangling, ggplot2 for graphs, and several more) that share a common style, plus a single meta-package of the same name that installs them together and loads the core ones for you in one step. It became enormously popular by smoothing everyday data work, and its pipe-based style — chaining steps with %>% — shaped how a lot of modern R is written (base R later added its own pipe, |>). jstats shares the goal of smoothing the path but takes a different route, staying close to base R.

jstats doesn’t reinvent everything from scratch — it stands on several well-regarded packages: ggplot2 for graphs, haven and labelled for reading and handling data (and its value labels) from SPSS, Stata, and SAS, readxl and writexl for Excel files, and rlang for some of the behind-the-scenes machinery. When you install jstats, all of these come along automatically and it uses them under the hood, so you never install or load them yourself. It’s a similar convenience to the way the tidyverse bundles its own family of packages — except jstats keeps these working quietly in the background rather than putting them in your hands to use directly.

You’ll also hear about R Markdown and its newer cousin Quarto — tools for weaving R code and writing together into reports, slides, and websites. These very guides are built with Quarto, with jstats commands and their output sitting right inside the pages you’re reading.

When you install R, and most packages, they come from CRAN — the Comprehensive R Archive Network, the official online library where R itself and thousands of packages are published and kept. It’s maintained by the R community and is free to use. (jstats isn’t on CRAN just yet — for now it installs from another common home for R projects, which the Install jstats page walks you through — but it’s headed there.)