Introduction to Structural Equation Modeling
HTML-код
- Опубликовано: 1 июн 2021
- Introduction to SEM seminar originally given on February 22, 2021.
This is the second seminar in a three-part series.
1. Confirmatory Factor Analysis (CFA) in R with lavaan
stats.idre.ucla.edu/r/seminar...
The first seminar introduces the confirmatory factor analysis model, and discusses model identification, degrees of freedom and model fit.
2. Introduction to Structural Equation Modeling (SEM) in R with lavaan
stats.idre.ucla.edu/r/seminar...
The second seminar explores structural equation models which is an umbrella term that encompass linear regression, multivariate regression, path analysis, CFA and structural regression.
3. Latent Growth Models (LGM) and Measurement Invariance with R in lavaan
stats.idre.ucla.edu/r/seminar...
The third seminar introduces latent growth modeling and how it relates to hierarchical linear models (HLM) and 2) measurement invariance in CFA and how to compare model fit between invariance models.
I swear to god, this is the least technical intro I've ever seen in a stat course. It speaks volumes of the teachin style. Amazing job.
Dr Lin you are an amazing educator. Your very engaging lecture really demystified SEM for me. Thank you so much. All the best
Dr. Lin thank you very much for your effort in explaining CFA and SEM so thoroughly theoretically and practically. Best seminars ever!
Clearly explained. Thank you Dr Lin
Thank you for this seminar! Brilliantly explained!
Dr. Lin is a fantastic instructor!
This is a great seminar. Thank you, Dr. Lin, for presenting a complex method in such a clear and understandable way!
One of the best explanations on RUclips, thank you for sharing!
This is wonderful!
Thank You Dr. Lin.
Thank you so much! 😊 The video clarified so many things and basically gave me a foundation to move forward!
That was an excelent class!
Thank you so much. This was very helpful. I can't believe how much you packed into such a small period of time. Well worth watching all the way through. :)
Great stuff! Thanks!
Thank you so much for publishing this seminar videos, very helpful!!
You’re welcome!
Exercices are so useful ; and one better understand if one can represent the same thing with different langages : matrices, conceptual diagrams, equations, output from R
Thank you, dear professor, super video I have a question: why estimate the variance of endogenous variables? Because the objective of a path analysis model is the estimates of three types of parameters: The paths, The covariances between the exogenous variables, and the variances of the exogenous variables. To determine the direct, indirect, and total effects between the variables. To avoid the Heywood cases, it is better to fix the variance of the endogenous variable to its empirical variance. And thus the parameters (Psi) variance of disturbance is constrained parameters not free?
Hello.
How is it possible we can fit the model
risk =~ verbal + ses + ppsych
if we don't have any value for the variable risk in the csv file
?
Hi, I have a basic question at 2:08:28, why the column of "Std.all" represents the loadings lambda instead of the column "estimate" ?
Thank you for such a detailed introduction to SEM.
I have a question - at 46:18 the model is called a "saturated model" because the df = 0. However I have been reading that a "saturated model" occurs when there are the same number of parameters as there are data points. In this case, the model has 5 parameters and 500 data points. Is it still a saturated model then?
how do we get the value of the latent ? not the variance but the value
Linear regression and multiple regression do not include the variance of endogenous variables. I could not understand why we use the variances. Would any of you help me?
Can SEM be built for Likert-type (ordered) data?
Could we use categorical variable in SEM?
I asked about dependant variable to be categorical