Written and maintained by CASRAI Editorial Board
Last updated
JASP is a free, open-source statistics program with a point-and-click interface, built for the analyses most researchers run every day. This guide covers what it is, how to install it from the official source, a menu-by-menu run through an independent-samples t-test, how to read the output, and which settings change what you report. Facts below come from the JASP project’s own site and documentation, checked 2026-10-29.
What JASP is used for in research
What is JASP?
JASP is a free, cross-platform statistics program with a graphical interface. It offers both classical (frequentist) and Bayesian versions of standard analyses, shows results in a panel that updates as you change the data or options, and lets you annotate the output. It is developed by the JASP Team and supported by the University of Amsterdam (jasp-stats.org). The peer-reviewed description is Love et al. (2019), JASP: Graphical Statistical Software for Common Statistical Designs, Journal of Statistical Software 88(2), which covers t tests, ANOVAs, linear regression and contingency tables, with Bayesian hypothesis testing and parameter estimation (doi:10.18637/jss.v088.i02).
What does JASP stand for?
JASP stands for Jeffreys’s Amazing Statistics Program, in recognition of the Bayesian pioneer Sir Harold Jeffreys (JASP FAQ).
Licence, engine and file formats
According to the FAQ, JASP is free and open source: the analysis engine is under the GNU GPL v2 and the desktop interface under the GNU AGPL v3. It uses R and CRAN packages for computation. It opens plain-text and spreadsheet files (.csv, .tsv, .txt, .ods, .xls, .xlsx) and data files from other packages: SPSS (.sav, .zsav, .por), SAS (.sas7bdat, .sas7bcat, .xpt), Stata (.dta), R (.rdata, .rds) and Minitab (.mwx, .mpx). That makes it a practical way to open a colleague’s SPSS or Stata dataset without owning that package. For how it sits next to its closest free rival, see jamovi vs JASP; for the commercial incumbent, see the SPSS guide.
Install or access (official source only)
Download only from the project’s download page. It lists these routes (checked 2026-10-29):
| System | Official install route |
|---|---|
| Windows | Microsoft Store (recommended; updates automatically), or the MSIX package (no automatic updates), or the MSI installer (uninstall older versions first), or the command winget install JASP.JASP |
| macOS | Separate builds for Apple Silicon and Intel |
| Linux | Flathub (Flatpak) |
| ChromeOS | Flatpak |
The page lists a 64-bit system, about 4 GB of disk space and at least 4 GB of RAM (8 GB recommended, 16 GB or more for large datasets). JASP runs without an internet connection. Which operating-system versions are currently supported, and the current release number, change over time, so read them from the download page rather than from a guide.
Step-by-step: run and report an independent-samples t-test
The walkthrough follows the worked example in Goss-Sampson, Statistical Analysis in JASP: A Guide for Students (7th edition, 2025, written for JASP v0.19.3, CC BY 4.0), which compares weight loss between two groups. Menu names can shift between releases, so use the in-app Help button if a label differs in your version.
- Open the data. Click the icon at the top left to open the main menu, choose Open, and pick a recent file, a file on your computer, the Open Science Framework, or the Data Library (50+ datasets). The guide’s example datasets are at osf.io/bx6uv.
- Check the variables. In the spreadsheet view, confirm each variable is set to the right type (nominal for groups, scale for measurements). The Data preferences hold the nominal/scale threshold and missing-value codes.
- Look at the data first. Run descriptives with distribution plots or boxplots by group before any test.
- Run the test. Go to T-Tests > Independent Samples t-test. Put the outcome (weight loss) in the Dependent variable box and the group variable (gender) in the Grouping Variable box.
- Tick the assumption checks. Request the Shapiro-Wilk test of normality and Levene’s test of equal variances. Tick Cohen’s d for the effect size.
- If Levene’s test is significant, report the Welch t statistic, its degrees of freedom and p instead of the standard Student result.
- If normality fails, tick Mann-Whitney in the same analysis, with Location parameter and Effect size. JASP then reports the rank-biserial correlation and the Hodges-Lehmann estimate.
- Save. Save as a .jasp file, which stores the data, annotations and analyses together. Export results to HTML or PDF, and data to .csv, .tsv or .txt. If the source file changes, Sync data (Ctrl-Y) refreshes it.
The same flow (menu, move variables into boxes, tick the options, read the live output) applies to the other tests. For the same analysis in a rival package, see the t-test in SPSS guide.
Output and how to read it
In the student guide’s example the output has four tables and three graphs. The parts you act on are:
- Assumption tables. Shapiro-Wilk and Levene’s test decide whether the Student, Welch or Mann-Whitney result is the one to report.
- Test table. Gives the t statistic, degrees of freedom and p value. In the guide’s example, t(85) = 6.16, p < .001, Cohen’s d = 1.322. These numbers belong to that guide’s sample data and show the reporting pattern: statistic with degrees of freedom, p, then effect size.
- Plots. Plots can be saved as images or sent to PowerPoint (.pptx).
Settings that change what you report
Open Preferences before you copy results into a manuscript:
- Results: choose exact p values (for example p = 0.00087) instead of p < .001, the number of decimals, plot resolution and a white or transparent plot background.
- Data: the spreadsheet editor, the nominal/scale threshold and missing-value codes.
- Interface: light or dark theme, language and zoom.
Tables copy out in APA format, and the JASP project notes that tables can also be copied as LaTeX. Syntax Mode shows the analysis as R code, which helps when you need to document what was run. For Bayesian output, see interpreting Bayes factors.
Troubleshooting
- Installing over an older MSI version. The download page says to uninstall older versions first.
- No automatic updates. The MSIX package does not update itself; the Microsoft Store version does.
- Source data changed. Use Sync data (Ctrl-Y).
- Unsure what an option does. Use the Help button in the analysis panel; it opens the help file for that analysis. Official learning resources (Getting Started, Data Library, Video Library) are on the JASP materials page.
Free alternatives
| Program | What the official site says | Source |
|---|---|---|
| jamovi | A free and open statistical spreadsheet, powered by R; runs in a browser (cloud) or as an offline desktop app. See the jamovi dictionary entry. | jamovi.org, checked 2026-10-29 |
| R | A language and environment for statistical computing and graphics, distributed as free software under the GNU GPL; extended through add-on packages; runs on Linux, Windows, macOS and other systems. | r-project.org, checked 2026-10-29 |
FAQ
Is JASP free?
Yes. The project describes it as a free, open-source program (engine under GNU GPL v2, interface under GNU AGPL v3).
What is JASP used for?
Common statistical designs: t tests, ANOVA, regression and contingency tables, in classical and Bayesian form (Love et al., 2019).
What does JASP stand for?
Jeffreys’s Amazing Statistics Program, after Sir Harold Jeffreys.
Does JASP run without internet?
Yes. The download page states no internet connection is needed to run it.
Can JASP open SPSS files?
Yes: .sav, .zsav and .por, along with SAS, Stata, R and Minitab files, per the JASP FAQ.
Sources
- JASP Team. JASP home page, FAQ, download page and how to use JASP. Date checked: 2026-10-29.
- Goss-Sampson, M. (2025). Statistical Analysis in JASP: A Guide for Students, 7th ed. PDF. Date checked: 2026-10-29.
- Love, J., Selker, R., Marsman, M., et al. (2019). JASP: Graphical Statistical Software for Common Statistical Designs. Journal of Statistical Software, 88(2), 1-17. doi:10.18637/jss.v088.i02.
