Open-source converters for meta-science

Convert statistical software files to reproducible R Markdown

Upload a Stata .do, JAMOVI .omv, JASP .jasp, or SPSS .sps file and get back an .Rmd + knitted HTML report you can read, edit, and re-run. Built for replication research and transparent open science.

4
Source formats
Stata, JAMOVI, JASP, SPSS
180+
Commands / analyses
across the four converters
1,200+
Automated tests
in the R packages combined
MIT
License
All packages are open source

Why this exists

A lot of published analyses live inside proprietary or point-and-click tools. Re-running them, auditing them, or pulling a workflow into a teaching example often means rewriting the analysis from scratch in R. 2Rmarkdown gives you a starting R Markdown document that mirrors the original analysis, so you can read it, knit it, modify it, and share it with collaborators.

What it covers

Four formats, one consistent output

Each format has its own converter, tuned to that ecosystem's quirks. Every converter produces the same kind of artifact: an.Rmdplus a knitted HTML report, with the original syntax preserved as comments.

Stata

R CMD check: 0 errors, 0 warnings, 0 notes

Deterministic converter for Stata .do syntax files. Parses 53+ commands, expands macros (local, global, foreach, forvalues), handles factor variables and clustered/robust SE, and generates self-contained R Markdown reports targeting the fixest + modelsummary + tidyverse ecosystem.

53+ Stata commands with dedicated converters

95.9% command coverage across 237 real .do files

715+ automated tests, all passing

R backend: fixest + modelsummary + tidyverse

JAMOVI

R CMD check: 0 errors, 0 warnings, 1 note

Extracts data and analysis configurations from JAMOVI .omv files, reconstructs the jmv R package function calls, and generates formatted R Markdown documents with all original analyses preserved. Supports descriptives, t-tests, ANOVA, correlation, regression, reliability, factor analysis, and more.

20+ jmv analysis functions supported

315 automated tests across 13 OMV files

Incomplete analysis filtering (skips empty formulas)

R backend: jmv

JASP

R CMD check: 0 errors, 0 warnings, 1 note

Extracts data and analysis configurations from .jasp archives (ZIP files containing SQLite databases and JSON metadata), maps them to equivalent R code using 80+ function mappings across 17 JASP packages, and generates formatted R Markdown documents. Covers frequentist and Bayesian analyses.

80+ function mappings across 17 JASP packages

Tested against 137+ .jasp files

Parse-only mode (avoids JASP package fragility)

R backend: jaspTools / standalone R

SPSS

R CMD check: 0 errors, 0 warnings, 1 note

Parses 30+ SPSS commands including full expression translation (17+ SPSS functions mapped to R equivalents), translates them to R code using the jmv package, and generates formatted R Markdown reports. Preserves variable labels and value labels from .sav files. Validated against GNU PSPP.

30+ SPSS analysis commands supported

17+ SPSS functions translated to R equivalents

246+ automated tests, all passing

R backend: jmv + tidyverse

How it works

Three steps from upload to report

01

Upload

Drop your source file (and the matching dataset, when the format needs one). The web UI auto-detects which converter to use.

02

Parse & generate

The matching R package parses the syntax or analysis spec, maps it to equivalent R code (fixest, jmv, tidyverse, …), and assembles a self-contained R Markdown document.

03

Render & download

The Rmd is knitted to HTML on the server. You download both files. The HTML contains the executed R code's actual output — not a copy of the source app's screen.

Open-source R packages

Use the web app, or run the same converters locally

The web app is a convenience layer. Every converter is also a standalone R package you can install with remotes::install_github() and call from your own scripts or pipelines.

# install.packages("remotes")
remotes::install_github("giladfeldman/STATA2Rmarkdown")
remotes::install_github("giladfeldman/jamovi2rmarkdown")
remotes::install_github("giladfeldman/jasp2rmarkdown")
remotes::install_github("giladfeldman/spss2rmarkdown")

All five packages (including the shared stat2rmarkdown core) are MIT licensed. Pull requests, bug reports, and additional test corpora are welcome.

What to expect

Honest about what this is

This is a working tool, not a finished product. The converters cover the analyses we've tested against real published scripts and replication corpora. Output is meant as a faithful starting point in R, not a guarantee of identical numerical results across every edge case. Always sanity-check the generated Rmd against the original analysis before relying on it.

Quotas & privacy

Free for researchers, modest limits

  • Sign in with Google or GitHub — no separate account.
  • Soft caps of 5 conversions / day and 50 / month per user.
  • Uploaded files are processed in isolated worker containers and deleted after the run.
  • Need a higher cap? Email giladfel@gmail.com.

Ready to convert your first file?

Sign in with Google or GitHub. Your first conversion takes about twenty seconds while the worker spins up.

Sign in to convert
vc000fee