| lavaan | |
|---|---|
| Original author | Yves Rosseel |
| Developer | Yves Rosseel (main developer) |
| Initial release | 11 May 2010 |
| Stable release | Version 0.6-21, Released on CRAN / 21 December 2025 |
| Written in | R |
| Operating system | Windows, Linux, macOS |
| Website | lavaan |
| Repository | github |
lavaan is an open-source structural equation modeling (SEM) package for the statistical programming language R. Created by Yves Rosseel in 2010, lavaan is intended to be a free open-source commercial-quality alternative to commercial (SEM) software packages such as Mplus or IBM SPSS Amos. [1] [2]
On 11 May 2010 lavaan version 0.3-1 was released on CRAN, the original release of lavaan. [3]
On 21 February 2011 lavaan version 0.4-7 was released on CRAN, launching the 0.4 series of releases. [4]
On 8 September 2012 lavaan version 0.5-9 was released on CRAN, launching the 0.5 series of releases. [5]
On 22 May 2018 lavaan version 0.6-1 was released on CRAN, launching the 0.6 series of releases. [6]
On 21 December 2025 lavaan version 0.6-21 was released on CRAN, the most recent release of lavaan. [7]
A variety of open-source, free, and commercial structural equation modeling software packages are available besides lavaan, such as Amos, SAS PROC CALIS, R package sem, OpenMx, LISREL, EQS, and Mplus. The main difference between packages is the presence of a graphical interface for model specification and presentation of results. Each package differs in terms of its strengths and weaknesses. lavaan was shown to be feature equivalent to commercial packages, with a compact scripting capability even for complex models as a key strength, and the ability to mimic Mplus and EQS results as a special feature. [8]
After 15 years of being developed, lavaan has become one of the standard structural equation modeling software package of choice for many researchers and data scientists across the world; evidenced in its use in studies, part of software reviews, and published books. [8] [9] [10] It was downloaded more than 4 million times to date. [11]