Solving a massive problem with scientific models: visualizing latent variables

[@Martyr022]

Looking forward to a YT series on SEM and Latent Variable Modeling! :)

[@TheBjjninja]

Dr. Fife, You have the best videos. Thank you.

[@luisa1551]

Yes! I would love to learn more about latent variables 😊

[@atinderbharaj]

dude, you're amazing. this is super interesting

[@LionsYouth]

Thank you for your efforts. Love the amalgamation. Can't wait to try out flexplavaan.

[@emilyandlilythesquishmallo969]

Like it very much. Thanks

[@groadoswaggins]

Quick question please. I see lmer @ 2:37. I have some mixed models (using lmer) where the main (plotted) random effect is categorical (i.e. the x-axis are the species (plotting means with CI), y axis are values). Will this work on that type of model? Would love to see a walkthrough if so. Love this stuff. Keep it up. Cheers.

[@QuantPsych]

It will work. Search my channel for mixed models and you'll see examples of that.

[@simonvrhovec1515]

Nice work! Let's say that we have a structural model with 4 latent variables (3 independent IVA, IVB and IVC, 1 dependent DV). I'm wondering whether is it possible to visualize relationships (structural plots) between individual independent variables and the dependent variable (e.g., just IVA and DV). For example, if we use the code "visualize(model.fit, plot="latent", method="lm", formula=DV~IVA)", it does not appear that the formula parameter does the job (the visualization includes all 4 variables). Am I missing something?

[@tatjanajak]

As many others, I look forwad to watch and listen about this new thing called Latent Variable Modeling. Never heard of it, and thank you for the introduction. I wonder to which degree LVM could be relevenat in machine learning.