# JASP Desktop: a free SPSS-style statistics package for Bayesian and Frequentist analysis > JASP is a cross-platform statistics application built around a point-and-click interface, with an R engine underneath for the analyses and QML for the dialogs. It suits researchers who want SPSS-familiar workflows without writing code. **jasp-stats/jasp-desktop** — JASP aims to be a complete statistical package for both Bayesian and Frequentist statistical methods, that is easy to use and familiar to users of SPSS - Repository: https://github.com/jasp-stats/jasp-desktop - Website: jasp-stats.org - Stars: 1,001 · Forks: 232 - Language: C++ - License: AGPL-3.0 - Published: 2026-08-08 · Updated: 2026-08-18 · Language: en - Canonical page: https://hysenlabs.com/projects/jasp-stats-jasp-desktop ## What JASP replaces, and for whom JASP targets a specific gap: analysts who need inferential statistics but do not want to write R, Python or SPSS syntax to get them. The README states the goal directly, describing a package that lets you "conduct statistical analyses in seconds, and without having to learn programming or risking a programming mistake." That second clause matters more than it looks. A mistyped formula in a script produces a wrong number silently; a dialog box with a fixed set of fields produces whatever the fields say. The intended audience is the SPSS user. The project says it aims to be "familiar to users of SPSS," which is a statement about layout, terminology and the shape of the output tables, not just about feature coverage. Someone who has used SPSS for a decade can open JASP and find the analysis they want in a menu rather than in a help page. The scope is deliberately dual: Bayesian and Frequentist methods in the same package. That combination is uncommon in free tools. Most either lean Frequentist with Bayesian added later, or the reverse. JASP treats both as first-class, which is why a single dialog can offer a classical t-test and a Bayes factor side by side. ## How the desktop application is put together The repository layout tells you most of the architecture. There is Desktop/, Engine/, QMLComponents/, R-Interface/, SyntaxInterface/, Modules/, Common/ and CommonData/. The split between Desktop and Engine is the important one: the interface and the computation are separate processes, which is why a long analysis does not freeze the window. QMLComponents/ and the QML guide under Docs/development confirm that the user interface is written in QML, Qt's declarative language. Each analysis gets a QML file describing its input controls. The R-Interface/ directory and the R analyses guide point to the other half: the actual statistics run in R, driven by a package list the project maintains separately. Modules/ is where analyses are packaged. The README links a tutorial, a module structure document and a module workflow document, which together describe how a third party adds an analysis to JASP without touching the core. That is the extension model: you write QML for the interface and R for the computation, and the module system wires them together. SyntaxInterface/ is worth noting because it undercuts the pure point-and-click story slightly. There is a syntax layer, so the menu choices can be exported as something scriptable. The README does not document its scope, so treat it as an area to investigate rather than a guaranteed R-compatible scripting environment. ## Installing JASP and running a first analysis The README does not give an install command. It points to a download page, and that is the supported path for most people. JASP ships as a desktop application for Windows, macOS and Linux, so installation is a download and run, not a package manager invocation. The same README section links an introductory guide on how to use JASP and a blog for feature announcements. If you want the newest build rather than a release, the README links nightly versions at static.jasp-stats.org/Nightlies/, described as built fresh every night. Those are development builds; the README does not describe their stability, so treat them as a way to preview changes rather than a deployment target. For a first real use, the workflow is: open the application, load a data file, pick an analysis from the menu, and assign variables to the fields in the resulting dialog. The output appears as you change the assignment, which is the behaviour that makes the tool feel fast compared to a script-run-inspect cycle. The README's own framing is that analyses take seconds. Building from source is a separate path with its own documentation. The README lists an introduction plus per-platform guides: ```bash # See Docs/development/jasp-build-guide-macos.md # See Docs/development/jasp-build-guide-linux.md # See Docs/development/jasp-build-guide-windows.md ``` Those guides are the authority on toolchain requirements. The repository root has a CMakeLists.txt and a conanfile.py, which indicates a CMake build with Conan handling dependencies. The README does not state the minimum compiler or Qt version, so read the platform guide before starting a build. ## Where JASP is the wrong tool The point-and-click model is a real constraint, not just a convenience. If your analysis needs a custom likelihood, a bespoke resampling scheme, or a model that no module implements, JASP gives you a dialog that does not exist. You would be writing R anyway, and at that point the interface is overhead rather than help. Reproducibility is the second issue. A GUI workflow is harder to put under version control than a script. The SyntaxInterface/ directory suggests the project is aware of this, but the README does not document how complete the exported syntax is or whether it round-trips. Anyone whose work must be re-run byte-for-byte by a reviewer should confirm that behaviour before standardising on JASP. Third, the module ecosystem is the ceiling. Analyses live in Modules/, and the README's development section is largely about writing more of them. A method that is not in a module is not in JASP. The project does not claim to cover every technique, and the R package list it links is a list of dependencies rather than a statement of analytical coverage. Finally, the licence. AGPL-3.0 is a strong copyleft licence with a network clause. For an individual researcher or a university department running the application, that is unremarkable. For a company that wants to embed JASP's engine inside a hosted product, the obligations are substantial. This is not legal advice; the repository's Docs/development/jasp-licensing.md is the place to read the project's own explanation. ## JASP against SPSS and against scripted R The comparison the README invites is with SPSS, and the difference is cost and licensing rather than capability. SPSS is commercial and its licence terms restrict who can install it and where. JASP is free to download and its source is public. For a student, a small lab, or a researcher in a country where a departmental SPSS licence is not available, that difference decides the question. The more interesting comparison is with scripted R. R gives you unlimited analytical reach and a text file you can commit to git. JASP gives you speed and a lower error rate on standard analyses, at the cost of reach and scriptability. These are not competing on the same axis, and the honest answer is that many people use both: JASP for the routine work, R for the parts JASP does not cover. The R-Interface/ directory means JASP is not hiding R from you, it is wrapping it. There is also a Bayesian angle. A frequentist-only free tool plus a separate Bayesian package is a common combination. JASP putting both behind the same interface means you can run a t-test and get a Bayes factor without switching programs, which changes how often people actually report both. ## Maintenance, releases and the cost of upgrading The repository is not archived, and the last push was on 2026-07-07. Recent releases are v0.98.1 on 2026-07-07, v0.98.0 on 2026-07-01 and v0.97.1 on 2026-06-05, so the project is issuing tagged versions on a roughly monthly cadence across that window. The default branch is development, which means the branch you land on by default is not a release branch. For an end user, upgrade cost is low: download the new installer and your data files remain readable. The README does not document a migration path or a file format version, so if you keep analysis files between versions, verify that they open in the new release before deleting the old one. For a module developer, the cost is higher. The README points to a module workflow document and a module structure document, and any change to the QML interface conventions or the R-side analysis contract ripples into third-party modules. The nightly builds exist precisely because the development branch moves. If you maintain a module, pin to a release tag rather than tracking development, and read the module workflow document before each major upgrade. The licence is AGPL-3.0. Redistributing a modified JASP, or offering it as a network service, brings source-disclosure obligations. The repository's own licensing document under Docs/development is the reference; nothing here substitutes for reading it. ## Conclusion Adopt JASP if your analysis workflow is menu-driven, your team is comfortable with SPSS conventions, and you want Bayesian and Frequentist output from the same dialog. Do not adopt it if you need a scripted, reproducible pipeline as the primary interface, or if AGPL-3.0 redistribution terms conflict with how you ship software. Before committing, verify that the analyses you need exist in the module set, and check the build guides under Docs/development if you plan to compile rather than download. ## FAQ ### How do I download JASP on my computer? The README points to the download page at jasp-stats.org/download/, which offers JASP for Windows, macOS and Linux. Installation is a download-and-run process rather than a package manager command. Nightly development builds are also linked from the README at static.jasp-stats.org/Nightlies/. ### What is better, SPSS or JASP? The README describes JASP as easy to use and familiar to users of SPSS, so the interface deliberately follows SPSS conventions. The practical difference is licensing and cost: SPSS is commercial, while JASP is free to download with public source under AGPL-3.0. The README does not make a feature-by-feature comparison between the two. ### Can I download JASP on a Mac? Yes. The README states that JASP can be downloaded for Windows, macOS and Linux, and there is a dedicated macOS build guide under Docs/development for compiling from source. ### Is JASP difficult to use? The README states that JASP lets you conduct statistical analyses in seconds without having to learn programming, and that it aims to be familiar to users of SPSS. The repository also links introductory guides on how to use JASP for new users. ## Sources - [Official README](https://github.com/jasp-stats/jasp-desktop#readme) - [Project repository](https://github.com/jasp-stats/jasp-desktop) - [Release notes](https://github.com/jasp-stats/jasp-desktop/releases) --- Hysen Labs editorial analysis, written from the project's own repository and release notes. Cite the canonical page: https://hysenlabs.com/projects/jasp-stats-jasp-desktop