About jstats
jstats makes statistical analysis in R approachable for social-science researchers and students. R is powerful, but its learning curve and rough edges can make the move from commercial statistical software like SPSS, Stata, or SAS intimidating. jstats smooths that path: shorter, more consistent syntax; sensible defaults; protection from some of base R's more confusing behaviors; and output styled after the conventions the expensive commercial platforms use, so results look familiar from the first run.
Take the hurdle that stops many newcomers in the first ten minutes: simply opening a data file. In jstats, once you've set a single option telling it which folder your dataset lives in, opening that file takes just one command — regardless of whether it came from SPSS, Stata, or Excel — and the value labels, variable labels, and missing-value codes those files carry come in with it, handled automatically and with no manual conversion. In base R the same task is several separate steps: pick and load the right package for that file type, read the data, then work out and declare which values should count as missing. jstats acts as a wrapper — it folds those steps into one function call, doing the busywork behind the scenes so you can get to the analysis. And there's plenty to get to: jstats covers the analyses social scientists reach for most — frequency tables and descriptive statistics, data screening, t-tests and ANOVA, correlation, scale reliability (Cronbach's alpha), cross-tabulations, and both linear and logistic regression — with a fuller list on the Reference page.
And for all the steps it folds together, jstats stays close enough to base R that the skills you build transfer to the wider R world rather than locking you into a private dialect — ordinary R is always right there underneath, ready when you need it. None of this was designed in the abstract: jstats grew out of years of teaching statistics to social-science students who wanted the methods, not a programming course, and which analyses it includes, which defaults it sets, and which rough edges it smooths all come from what real students and colleagues run into when they put it to work.
About the developer
I'm Jeff Ackerman, a social scientist who teaches quantitative methods at Griffith University in Queensland, Australia. I grew up in Pennsylvania, in the United States, and joined Griffith in 2013; I'm now an Australian citizen. I studied statistics and research methods at the Pennsylvania State University under several well-known methodologists, including C. R. Rao — himself a student of Ronald Fisher, the statistician who developed Analysis of Variance (ANOVA). Alongside statistics, I trained in computer programming at the university level decades ago, and that combination — now paired with modern AI-assisted software engineering and careful testing — is what makes it practical for one person to build and maintain a package this size.
New here?
The Get Started guide takes you from nothing installed to your first real statistical result — an actual regression — in about an hour, with no prior R experience needed. You’ll do a little setup once, then spend your time getting results.