Updated 10/23/2026
Errors when you use free statistical software for Macs?
See our “signing page.”
JASP, Jamovi, and PSPP are the free Swiss Army knives of Mac statistics software. They are all faster than SPSS, and import and export SPSS data and syntax files.
PSPP was first, but development has been slowest, and it is rather behind the others now. JASP was forked from Jamovi, but both are still under active development. They are interfaces for the well proven R software.
See a comparison of the three SPSS competitors
All three major SPSS replacements have some oddities as well as major strengths, and notably all produce the same numbers when we test them.
Listing updated 10/23/26
Versions: Homebrew, 2.12; source, 2.13*; binary, 2.12
Download size: quite small!
* (Homebrew update tends to come days or weeks after source updates, then binary)
PSPP is a free SPSS clone with a Mac version you can download from this site (it’s signed and notarized). You can also build from source with HomeBrew or MacPorts. It is small and loads quickly, but has bugs on Macs and Windows—it prefers Linux. Development is slow but continues. Here’s a video on installing it.
PSPP imports SPSS data files, complete with long variable names and variable and value labels. As you run analyses, it appends them to the output window, instead of overwriting them if you change things. This can be very handy for comparisons.
Our getting-started guide for SPSS is also helpful for getting started with PSPP.
PSPP’s look and feel is from SPSS of the 1990s. The program is impressive, including its graphing. The routines are well fleshed out. The user interface is very fast, and if you have an old (but not recent) SPSS output file, PSPP can probably open it.
Problems: PSPP doesn’t use the Mac open/save box, which makes using Dropbox, iCloud, etc. awkward. You can add new locations to the sidebar, which helps, but they have to go near the bottom, while Music, Pictures, and Movies are unaccountably and weirdly stuck in the sidebar. (Also, I was only able to add one shortcut to the sidebar, which is less than ideal.)
You can only copy plain text, formatted with spaces and pipes, from the output window. Regression does not allow any sort of multiple step entry.
Version 2.13 added bug fixes and new mean and covariance extensions to MATRIX.
If you've installed via Homebrew, upgrade from the Mac terminal by typing brew update and then brew upgrade when Update is done. Type psppire to load it. The binary acts like any other Mac program, but the binary isn’t updated as quickly as the Homebrew.
Bugs and problems (last updated 10-20-2026).
Listing updated: 10/20/2026
Version: 0.98.8
1.4 GB download, 2.2GB on disk!
JASP, from the University of Amsterdam, is fast and easy to use. It has a Mac-native open/save box (click Browse in their menu), and has value and variable labels and long description fields. It imports from SPSS, Minitab, Stata, and spreadsheets; and now has an AI, which we haven’t tested because it’s an AI, and interactive plots with the ability to add reference lines and points.
Since data may be sensitive, JASP can encrypt your data/work files. If your data is sensitive, also use a local AI model rather than the network.
JASP feels native to the Mac, and is far faster than SPSS. Its Bayesian statistics may confuse some people. Stepwise, forward, backward, and manual (“enter”) regression are all supported, almost unique among free software; it also includes MNLFA and other SEM analysis is also included.
You can assign value labels variable by variable only, without syntax, and variable labels are not always used in the results. Computing new variables is relatively easy now, but if you really want to make that (or recoding) easier, or if you want to figure out a way to add variable or value labels more easily, you can unzip the jasp file, then use sqlite (or phpLiteAdmin*) to make changes or move columns into a spreadsheet such as Libre/Open Office Calc or Excel. Doing the following from Terminal will provide a pipe-separated text file:
unzip yourfile.jasp sqlite3 internal.sqlite "select name,title from columns;" > columns.txt
* See instructions for using a local copy of phpLiteAdmin
Developers are responsive, and responded to my problems with a large survey file by confirm ing and fixing issues with large file handling and buggy files.
The main downside is the interface for recoding data. To get around that, the writers let you “synchronize”—export the data, change it, then bring it back, seamlessly. There are caveats: you define the native editing program, and it has to handle the format you imported. If you imported from SPSS format, it will export the data as SPSS format; you can't synchronize with data in a format from a program you don't have.
New columns (variables) can be created by inserting a blank column, then copy-pasting. That won’t move labels and custom missing values over, though. (Missing values can be defined program-wide or by variable, quite easily.)
When you do t-tests, if equal variances are not present, it prints out a warning, and you must switch to Welch’s t-test; it does not apply Welch’s t-test automatically, not that any program we’ve looked at does this (SPSS does use both tests every time, which adds some work for the user). The format for multiple t-tests is quite neat, though, and a recent version improved the visibility of the warning quite a bit.
t-tests require two-value group variables—you can't pick, say, 2 and 3 bathroom houses; you have to create a new variable consisting solely of 2 and 3 bathroom houses, the rest being hidden (or hide all the unwanted choices). This is a bit of a nusiance.
Also, variable lists don't show labels, so variable names have to be chosen carefully. This is still probably a good option, because long variable names wouldn’t fit in the list.
You can set the resolution of charts, so you can copy them at 300 dpi if you want.
JASP’s advantages over Jamovi include supporting forward, backward, and stepwise regression; Jamovi only supports "Enter” (though it allows multi-step. PSPP allows none of these.) JASP has nifty menus and nicer output; an R syntax plugin; and Bayesian statistics everywhere. Jamovi has its own advantages...see below.
For some documentation, see Learning Statistics with JASP.
A Machine Learning module has predictive statistics and visual modeling.
In our run-throughs, the numbers were identical to SPSS, PSPP, and Jamovi.
Dive more deeply into JASP (full MacStats review). Also see this video on using JASP to do a t-test and contingency tables when using a grouping variable with nine values... as an alternative to creating a new variable with just two.
Listing updated: 10/20/2026; updates are roughly once a month
See the an online “cloud” version
Signed by Apple
378MB download, 800MB in place
Jamovi is another open source package built atop R which has a spreadsheet interface, full graphics, and both syntax and menus. Your data, analyses, and options are saved in a single file, so others can reproduce your work.
The results are attractive (see above), with menus and many easy to reach options; this is probably a better interface than SPSS, since it requires far less button-clicking. Right-click on a section of output to copy it, and you can paste it into Word as a nicely formatted table or into BBEdit as plain text, formatted with spaces. There are three built in plot themes.
The online (cloud) version is responsive and free.
Syntax mode shows the underlying R commands, helping you to learn R or to make scripts to reproduce the same actions over and over. Data can be imported from SPSS, SAS, Stata, and spreadsheets. Variable labels are supported; value labels are not. In the past, not checked recently, export from Jamovi to SPSS resulted in errors on some data files as the number of characters in some fields was not correctly marked.
Jamovi has its own file open/close system but a click on "Browse" provides access to the native file selection system.
Jamovi’s menus are within its own window instead of at the top of the screen. Variable names should be cleaned before importing data, because the variable lists work in alphabetical order and don't show labels.
Linear regression only allows Enter but provides multiple blocks and easy entry of factors and weights.
The t-test issues (using a warning instead of the alternate test, requiring two-value variables) are the same as in JASP.
Other issues include no support for date or time variables; and no ability to direct output to new variables. You can’t copy and paste from BBedit into Jamovi, but must use its clunky data transformation system. Both JASP and Jamovi are enormous on disk, due to the integrated software — for Jamovi, R, Electron, Mantle, Python, and ReactiveCoca.
Other notes. Developer Jonathon Love pointed us to the Jamovi library of extras. A Jamovi blog post goes over its graphics. Numbers are unsurprisingly identical to SPSS, PSPP, and JASP in our testing.
Dive more deeply into Jamovi (full MacStats review).
Listing updated: September 2026 (last software update, October 25)
Signed by Apple (App store version) • The University of Oslo
“Past is free software for scientific data analysis, with functions for data manipulation, plotting, univariate and multivariate statistics, ecological analysis, time series and spatial analysis, morphometrics and stratigraphy.” That said, Dennis Helsel wrote, “While its name shows its origin (Paleontology), it is a full-fledged stat package which includes multivariate and permutation tests, with a nice interface.” There is good support for geographical and map-based statistics.
When Dennis says “full-fledged,” he isn’t kidding — the range of this software is stunning. Yet, the download is a mere 10 MB — far, far, far less than many others. What’s more, every new version brings a wide range of new features—even 0.0x versions. Summary statistics came in a fraction of a second on a laptop. Our survey file never caused more than a slight pause.
Downsides. Import formats are limited and exclude SPSS, Stata, and Excel files (checked with Past 5); some esoteric formats are accepted, though, and you can copy and paste from Excel (with caution). There are no value labels, but there are variable labels (click Column Attributes and you can enter variable names) and there is scripting. The Mac version may be a version behind Windows (Windows is at 5.3, Mac is at 5.2) and only supports M series processors now.
You can transform data but without the flexibility of some other software; you cannot set missing values. Transformations tend to the complex, leaving out the simple. The "select data and then operate on it” format is clunky; choose data groups instead of columns. There is no way to do independent-sample t-tests by having one variable define groups; you need to have each group in a different column, and only select the data you want to analyze.
Overall, Past is excellent for some of its features but people who don’t need its unique properties are better off with the others listed here.
Dive more deeply into PAST (full MacStats review).
Listing updated: 10/25/2024 • Size: 3.5 GB
SageMath is general math software that has some statistics. You can tap it from the command line or from Safari, Firefox, etc; and it can be put onto a server and shared. SageMath creates embedded graphics and typset-style math expressions; it is built atop NumPy, SciPy, matplotlib, Sympy, Maxima, GAP, FLINT, and R.
We only rarely recheck these listings.
Copyright © 2005-2026 Zatz LLC. All rights reserved. Created in 1996 by Dr. Joel West; maintained since 2005 by Dr. David Zatz. Contact us. Terms/Privacy. Books by the MacStats maintainer