[R] [R-pkgs] New package: `lavaan' for latent variable analysis (including structural equation modeling)

[Michael Friendly]

Hi Yves

lavaan looks like a very nice package.  From the tutorial introduction
I see you create path diagrams for some of the models you describe.
How did you do this?  I don't see a function for this in the package.

I know there is a path.diagram function in the sem package that uses
dot to draw the diagram, but I've always found the layouts from dot
somewhat strange for path diagrams without a fair bit of tweaking.

-Michael



Yves Rosseel wrote:
> Dear R-users,
> 
> A new package called `lavaan' (for latent variable analysis) has been 
> uploaded to CRAN. The current version of lavaan (0.3-1) can be used for 
> path analysis, confirmatory factor analysis, structural equation 
> modeling, and growth curve modeling.
> 
> More information can be found on the website: http://lavaan.org
> 
> Some notable features of lavaan:
> 
> - the 'lavaan model syntax' allows users to express their models in a 
> compact, elegant and useR-friendly way
> 
> - lavaan is robust and reliable: there are no convergence problems and 
> numerical results are very close (if not identical) to the commercial 
> package Mplus
> 
> - many different estimators are available: ML, GLS, WLS, robust ML using 
> Satorra-Bentler corrections, and FIML for data with missing values
> 
> - full support for meanstructures and multiple groups
> 
> - user friendly output including standardized solutions, fit measures, 
> modification indices and more
> 
> To get a first impression of how the 'lavaan model syntax' looks like, 
> below is the full R code for fitting a SEM model:
> 
> ## begin R Code ##
> 
> library(lavaan)
> 
> # The industrialization and Political Democracy Example
> # Bollen (1989), page 332
> 
> model <- '
>   # latent variable definitions
>      ind60 =~ x1 + x2 + x3
>      dem60 =~ y1 + y2 + y3 + y4
>      dem65 =~ y5 + y6 + y7 + y8
> 
>   # regressions
>     dem60 ~ ind60
>     dem65 ~ ind60 + dem60
> 
>   # residual correlations
>     y1 ~~ y5
>     y2 ~~ y4 + y6
>     y3 ~~ y7
>     y4 ~~ y8
>     y6 ~~ y8
> '
> 
> fit <- sem(model, data=PoliticalDemocracy)
> summary(fit, fit.measures=TRUE)
> 
> ## end R code ##
> 
> 
> Please feel free to contact me directly with questions and comments.
> 
> Best,
> 
> Yves Rosseel.
> 
> 
>