statlingo
Explain statistical model output with LLMs, in R and Python
statlingo translates the dense output of statistical models—coefficients, p-values, model fit indices, and more—into clear, context-aware, natural language explanations using Large Language Models (LLMs).
It ships as two packages, sharing a common design and a single canonical set of LLM prompts:
| Package | Language | LLM interface | Get started |
|---|---|---|---|
statlingo |
R | ellmer |
R quickstart |
statlingo |
Python | chatlas |
Python quickstart |
Why statlingo?
Statistical models are powerful, but their outputs can be intimidating. statlingo empowers you to:
- Democratize understanding — make complex analyses accessible to people with varying levels of statistical expertise.
- Accelerate learning — students can connect statistical theory to practical model output, with explanations tailored to their level.
- Streamline reporting — quickly draft plain-language interpretations for reports and presentations.
- Work in either language — the same audience/verbosity/style options and prompt engineering are shared across the R and Python packages.
Supported models
Both packages support a growing set of common statistical models — see the R function reference and Python API reference for the full, current list per language.
Source code
statlingo is developed in the open on GitHub: github.com/bgreenwell/statlingo.