Talk:Structural equation modeling

Suggestions from the Membership of the SEM-Net listserv

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)

Sample size section

"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

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)

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)

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)

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)

Hey MrOllie (talk · contribs), I had a look, and the German WP:NOT is pretty much the same as the English one, and there, the Table of Software is allowed. I don't see which rule this should break. You claim the link directory, but reading the text, I disagree that this is obviously the case. For me, the German software section is clearly more useful than the current version of the English one, which also links to a > 10-year-old (thus outdated overview), and I think many users would profit from the table. — Preceding unsigned comment added by Jkarch (talk • contribs)

It's obviously a link directory and is plainly not allowed per WP:NOT and WP:ELNO. If the German wiki has the same standards, it is likely that their article has improper links. That is not a reason improper links should be added here as well. - MrOllie (talk) 18:32, 18 January 2025 (UTC)

Please don't remove the table until a consensus has been reached. It's disruptive, and frankly, I do not see why it should be removed. I've already explained that this isn't a list of the purpose of advertising. It's a list to help users understand what tools are available. If your problem is with the links, then you can turn those into citations instead. ^SMason_Garrison 01:43, 19 January 2025 (UTC)

I raised this at Wikipedia:External_links/Noticeboard#List_of_external_links_at_Structural_equation_modeling. MrOllie (talk) 03:21, 19 January 2025 (UTC)

Also pinging @Mikeblas:, who removed this a couple years ago, leading to the recent stable version with no software list. - MrOllie (talk) 03:55, 19 January 2025 (UTC)

That still doesn't justify removing the content while the discussion was ongoing. You still haven't commented on the fact that this isn't recommendating specific software or my suggestion to convert the url to a citation. Further, the table reviews what capabilities the software has, which again isn't a link directory.WP:ELLIST ^SMason_Garrison 04:29, 19 January 2025 (UTC)

Remformatting an external link as a citation doesn't make much of a difference. This talk section had been idle for years, nothing was removed during an ongoing discussion. MrOllie (talk) 04:33, 19 January 2025 (UTC)

Discuss Differences to Bayesian Networks

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