JASP is a free, open-source statistics application that runs classical and Bayesian analyses side by side in one live interface, built for students, researchers, and teachers who want readable output without writing R code.
JASP is a statistics package built for people who read results rather than write code. You open a spreadsheet, choose an analysis from a menu, and fill in the variables; the output updates the moment an option changes, so a t-test, an ANOVA, or a regression can be adjusted until the specification matches the research question. What sets JASP apart is that every common analysis ships in two forms, a classical version reporting p-values and effect sizes and a Bayesian counterpart reporting Bayes factors, credible intervals, and posterior plots. Both appear in the same window, which makes the two philosophies easy to compare. Files are saved as .jasp projects that carry the data and analyses together, and CSV, Excel, and SPSS files import directly.
JASP grew out of an academic concern with reproducibility, and the interface reflects it. Because analyses are stored as settings rather than pasted output, reopening a project restores exactly which variables, options, and priors produced each table, and colleagues can inspect the choices instead of trusting a screenshot. Assumption checks, effect sizes, and robustness options sit close to the main result rather than buried in a separate menu. The module system extends the core install with specialised procedures, so a meta-analysis or a structural equation model arrives without switching programs. Output can be exported to HTML, PDF, LaTeX, or copied into a word processor, and analysts who need heavy time-series work often move the data across to gretl.
The practical gain of JASP is that it removes the argument about which statistics to report. A researcher can run a classical t-test and its Bayesian equivalent in adjacent panels, see whether both point the same way, and decide what the evidence supports before writing a sentence. Priors are exposed rather than hidden, so a sceptical reader can see the assumptions behind a Bayes factor and test how sensitive the conclusion is to changing them. Because JASP is free and open source, a lab, a seminar, or a whole department can standardise on it without licence negotiations, and students can install it on their own machines. The interface also lowers the cost of switching methods: the same dialog family covers t-tests, ANOVA, regression, contingency tables, and their Bayesian versions, so learning one analysis teaches you most of the layout for the next.
Key capabilities and highlights.
Every standard analysis has two entries in the menu, one reporting p-values, effect sizes, and confidence intervals, the other reporting Bayes factors, credible intervals, and posterior distributions. Running both takes seconds, so a results section can present the two perspectives without moving data between programs and reviewers can judge whether the conclusions agree.
Results redraw as soon as a variable, option, or prior changes. There is no compile step and no stale table left on screen from an earlier specification, which means exploratory comparison of model variants stays honest about what was actually run and no result is copied forward by accident.
Normality tests, homogeneity checks, outlier flags, and collinearity diagnostics sit beside the main table rather than behind a separate command. When an assumption fails, JASP shows which alternatives the data allow, so switching to a non-parametric or robust version is one menu selection away.
Data sits in a grid where variable types and value labels are declared once and applied everywhere. Filters, computed columns, and recodes can be layered on top, and the .jasp project keeps the original values intact so any reported figure can be traced back to the observation behind it.
The core install covers the everyday tests, while optional modules add meta-analysis, summary-statistics work, network analysis, structural equation modeling, and mixed models. Each module appears as an ordinary menu entry and its results behave like any built-in analysis, so the interface stays familiar when the method is specialised.
Rather than one fixed default, JASP exposes the prior distribution behind each Bayesian test and offers alternatives together with sensitivity plots. A reader who disagrees with your assumptions can see exactly what was specified and how far the conclusion shifts when the prior is changed.
Tables export to HTML, PDF, LaTeX, or a word processor, and figures are generated from the same results, so numbers and plots never drift apart. Styling stays consistent across analyses, which shortens the gap between finished output and a submittable manuscript. Copy-and-paste keeps captions and labels attached.
Because every analysis records its settings, an instructor can distribute a project file and have students reproduce each table exactly, then change a single option to see the effect. Effect sizes and assumption results are shown by default, which keeps the reasoning visible instead of hidden behind a decision rule.
Browse historical releases and previous builds.
Common questions about JASP.
Yes. JASP is free and open-source software, so students, teachers, and research groups can install it and use every built-in analysis without a licence fee, and the source code is available for inspection and reuse.
It does. Most standard procedures appear twice in the menu, once as a classical test reporting p-values and effect sizes and once as a Bayesian analysis reporting Bayes factors, credible intervals, and posterior plots, using the same loaded data.
No. Every analysis is driven through menus and option panels. Knowing R helps only if you want to understand the engine underneath or build your own module for a method the existing analyses do not cover.
It opens its own .jasp project files and imports common formats such as CSV, Excel workbooks, and SPSS data files, so an existing dataset generally needs no conversion before you can start specifying analyses.
A .jasp file is the native project format. It holds the dataset together with every analysis you built from it, including the options and priors you chose, so reopening it restores the exact tables rather than a blank workspace.
Yes. JASP is module-based, and additional modules covering areas such as meta-analysis, network analysis, structural equation modeling, and mixed models can be added and then used like any built-in menu entry.
Tables and figures can be exported to HTML, PDF, or LaTeX, or copied straight into a word processor. Because figures are generated from the same results as the tables, the numbers and the plots stay consistent through editing.
It suits teaching because outputs appear immediately as options change and effect sizes plus assumption checks are visible by default. Students can also be given a JASP project file to reproduce exactly, which makes a methods exercise verifiable rather than a matter of trust.
JASP is a free, open-source statistics application that runs classical and Bayesian analyses side by side in one live interface, built for students, researchers, and teachers who want readable output without writing R code.
JASP is a free, open-source statistics application that runs classical and Bayesian analyses side by side in one live interface, aimed at students, researchers, and teachers who want readable output without writing R code.
Minitab Statistical Software 22 is a commercial statistics package built around quality improvement, combining guided analyses, control charts, process capability, and design of experiments so engineers, analysts, and students can turn process data into decisions without programming.
gretl is a free, open-source econometrics package that pairs a point-and-click interface with its own scripting language for regression, time-series, and panel-data work, aimed at students, teachers, and researchers who want transparent, reproducible estimation.
Jamovi Desktop is a free, open-source statistics application with a spreadsheet-style workspace where analyses update live as you change variables, built for students, researchers, and analysts who need reliable results without writing code.
Comments