Talk:Structural equation modeling

Latest comment: 1 year ago by Biggerj1 in topic Discuss Differences to Bayesian Networks

Suggestions from the Membership of the SEM-Net listserv

edit

SEM-Net is a large and longstanding email listserv of people who are interested in Structural Equation Modeling. A group of them, led by Leslie Hayduk, put together a set of suggestions for content for the article. They do not have a lot of experience editing Wikipedia, so they gathered the material and emailed it to me to ask for help converting it to Wiki Markup. I am nearly done with this in my Sandbox, and I will start moving content over to here (Talk page) or to the main article. Suggestions or help moving content both welcome! Prof. Eric A. Youngstrom (talk) 18:36, 15 March 2022 (UTC)Reply


Sample size section

edit

"Complexities which increase information demands in structural model estimation increase with the number of potential combinations of latent variables; while the information supplied for estimation increases with the number of measured parameters times the number of observations in the sample size – both are non-linear. Sample size in SEM can be computed through two methods: the first as a function of the ratio of indicator variables to latent variables, and the second as a function of minimum effect, power and significance. Software and methods for computing both have been developed by Westland (2010)."

There is a long history addressing the issues of sample size in SEM. I have edited out what I feel are personal, unfounded and unscientific assertions from an anonymous poster. Indeed this quote is from my research: both the ECRA article and my Springer book "Structural Equation Models: From Path to Network Analysis". (JC Westland)

Software

edit

Hey MrOllie (talk · contribs), I don't think that directing folks to the software implementations of the method counts as a link directory WP:LINKSTOAVOID. [1]

>Researchers using SEM employ software programs (such as Mplus, lavaan (in R), LISREL, OpenMx, SPSS AMOS, Stata)

Would you be ok if we changed the external links to bluelinks?

> Researchers using SEM employ software programs (such as Mplus, lavaan (in R_(programming_language)), LISREL, OpenMx, SPSS AMOS, Stata)

I think that way we wouldn't lose the important information about software without feeling like we're advertising, which I think is your concern

Mason (talk) 12:29, 7 July 2022 (UTC)Reply

Making a list of external links to software is exactly a link directory, and exactly the sort of thing we shouldn't be doing. I also don't agree that lists of examples should be present. I don't feel this is 'important information' since Wikipedia isn't supposed to be a site that recommends particular software to users. - MrOllie (talk) 12:35, 7 July 2022 (UTC)Reply
Ah, ok. I can see where you're coming from. Let me think about it because as someone who teaches SEM, these are pretty much "all" the major software implementations that researchers use. I've found that folks are much more receptive to applying techniques if they're already familiar with the software. Mason (talk) 12:45, 7 July 2022 (UTC)Reply
Looks like we both missed the section on software, lol. Structural_equation_modeling#SEM-specific_software. I'll putt the missing information there.Mason (talk) 12:58, 7 July 2022 (UTC)Reply

Discuss Differences to Bayesian Networks

edit

E.g. cyclic dependencies are allowed. Could you add more, including literature on it? Biggerj1 (talk) 21:08, 22 May 2023 (UTC)Reply