In general Path Analysis can be considered as a special case of SEM - structural equation modeling.

You can find an introduction to Path Analysis, and examples in R here:

http://www.unt.edu/rss/class/mike/6810/Path.pdf

Here the author use the R package SEM.

Other packages which perform SEM modeling (and path analysis of course) are lavaan, openMX and SEM. A interesting tutorial on lavaan (which many examples of path analyses) is here:

http://users.ugent.be/~yrosseel/lavaan/lavaanIntroduction.pdf

The Partial Least Squares approach to SEM is the PLS Path Modeling which can be used in R by considering the package plspm. Here the guide:

http://gastonsanchez.com/PLS_Path_Modeling_with_R.pdf

Of course SEM (Covariance Structure Analysis\Lisrel approach) and PLS Path Modeling are different approaches with different underlying concepts behind.

I hope you can find useful this material,

Kind Regards,

Carlo Drago

following are the pathway based R packages :

http://bioconductor.org/packages/release/BiocViews.html#___Pathways

though listed in the previous link I am adding another link that contains documentation, example, tutorial....for Pathview package

http://pathview.r-forge.r-project.org/

Hi Amin,

For general introductions to R see the question here:

https://www.researchgate.net/post/What_resources_or_books_free_or_otherwise_do_you_use_for_R

For path analysis you can have a look at the SEM tools in the psychometrics task view

http://cran.r-project.org/web/views/Psychometrics.html

I would recommend lavaan as it is fairly simple to work with. Just follow the tutorials on the lavaan website:

http://lavaan.ugent.be/

In general Path Analysis can be considered as a special case of SEM - structural equation modeling.

You can find an introduction to Path Analysis, and examples in R here:

http://www.unt.edu/rss/class/mike/6810/Path.pdf

Here the author use the R package SEM.

Other packages which perform SEM modeling (and path analysis of course) are lavaan, openMX and SEM. A interesting tutorial on lavaan (which many examples of path analyses) is here:

http://users.ugent.be/~yrosseel/lavaan/lavaanIntroduction.pdf

The Partial Least Squares approach to SEM is the PLS Path Modeling which can be used in R by considering the package plspm. Here the guide:

http://gastonsanchez.com/PLS_Path_Modeling_with_R.pdf

Of course SEM (Covariance Structure Analysis\Lisrel approach) and PLS Path Modeling are different approaches with different underlying concepts behind.

I hope you can find useful this material,

Kind Regards,

Carlo Drago

Hi Amin, this doesn't directly address your question about R, but I want to point out that there are older approaches to path analysis based on ordinary least squares regression, not SEM. Surprisingly, some people still publish such models. Even though the SEM models are more difficult and may require larger sample sizes and may be hard to get the high fit statistics that seem to be required in the field, I don't recommend resorting to the older techniques as they face a lot of limitations. Bob

Thank you all for the answers and suggestions

It is easy  to use RAMpath R packages,you can download the software  frome the following website     http://cran.r-project.org/web/packages/RAMpath/index.html

I am using the package 'plspm'. I thought it was very straightforwards and it produces nice graphs.

I go with Kerstin Zander. plspm is very easy to use and has many features

Also check out the generalized Structural Equation Modeling (gSEM) package: https://cran.r-project.org/web/packages/gSEM/index.html

Based on Robert contributions, what are the limitations of path analysis using ordinary least squares regression against SEM

Thanks

Thank you for all answers and suggestions

'lavaan' package is a good option to conduct path analysis in R. There is a very good website where you can read about the package and about the path analysis: http://lavaan.ugent.be/tutorial/index.html

You can find tutorials, exemples, how to fit models,...