Authors: Jan de Leeuw (University of California, Los Angeles, USA) and Miodrag Lovric (Radford University, USA)
This article traces the historical evolution and current landscape of statistical software, emphasizing its central role in the development of modern data analysis. The journey begins with mainframe-based packages like BMDP, SAS, and SPSS, then transitions through personal-computer-centric tools such as Data Desk, JMP, and Stata, and culminates in the rise of programming-based platforms like S, R, and Python.
Early software—BMDP, SPSS, SAS—was built for mainframes and tailored to specific disciplines like health sciences and social research. Later, graphical and interactive systems such as JMP and Data Desk highlighted the importance of visual exploration. Stata stood out for its command-line interface and user-extensibility, supporting a dynamic user community.
A major transformation came with the development of the S language at Bell Labs, which led to S-plus and, ultimately, to the open-source powerhouse R. R’s ecosystem, including RStudio and CRAN, now dominates statistical computing in both academia and applied research. R allows flexible, extensible modeling, rapid prototyping, and reproducible reporting via tools like R Markdown, bookdown, and Quarto.
The article explores innovations like LISP-STAT, early Bayesian platforms, and the emergence of object-oriented graphics and scripting in R. It also covers Bayesian computation advancements using JAGS, Stan, PyMC3, and software-specific tools like SAS PROC MCMC and Stata’s bayes suite.
Central to the discussion is the issue of reproducibility and the growth of open-source culture. With CRAN offering over 20,000 R packages, and platforms like GitHub, RPubs, and arXiv supporting open science, software has become both a technical and philosophical pillar of statistical practice.
The authors conclude by highlighting challenges such as fragmentation in R’s design and scalability concerns, yet reaffirm the liberating power of statistical software to democratize analysis and expand scientific reach.
For detailed timelines, software comparisons, and references, consult the full article in the International Encyclopedia of Statistical Science.