Hi
I teach statistics to Psychology students at the University of Costa Rica. JAMOVI is a wonderful tool for us. I want to use JAMOVI for SEM in an upcoming course.
I would like to ask: Is there a way to get in the confirmatory factor analysis module the Mardía index, the Satorra-Bentler Chi-Square, the robust CFI and robust RMSEA, as well as, the Robust Errors? If it is not possible: Have you considered including them?
On the other hand, I doubt if the Structural Equation Modeling module when selecting the MLR method (Robust MAximun Likelihoos) JAMOVI offers me the global adjustment indexes (chi-square, CFI, and robust RMSE).
Thank you very much
Tomás
SEM robust options
-
mcfanda@gmail.com
- Posts: 462
- Joined: Thu Mar 23, 2017 9:24 pm
Re: SEM robust options
Hi
as regards the SEM, it is not clear what seem to be the issue. When you select MLR, you get the scaled tests (together with the unscaled) and the scaled and robust versions of the (majority of) indices. Do you think they are not correct or do you wish to have different tests?
as regards the SEM, it is not clear what seem to be the issue. When you select MLR, you get the scaled tests (together with the unscaled) and the scaled and robust versions of the (majority of) indices. Do you think they are not correct or do you wish to have different tests?
Re: SEM robust options
Hi
I apologize if my English was not precise enough. I was confirming that the MLR option's SEM module offered robust versions. With your answer, it is clear to me that it does.
In the case of confirmatory factor analysis, is there any way to get the robust fit indices?
Thank you very much
Tomás
I apologize if my English was not precise enough. I was confirming that the MLR option's SEM module offered robust versions. With your answer, it is clear to me that it does.
In the case of confirmatory factor analysis, is there any way to get the robust fit indices?
Thank you very much
Tomás
-
mcfanda@gmail.com
- Posts: 462
- Joined: Thu Mar 23, 2017 9:24 pm
Re: SEM robust options
Not at the moment, but you can perform a CFA as a SEM
Re: SEM robust options
Hi
I have been using the SEM module for robust fit indices. Congratulations and many thanks for your work!
I have not found detailed documentation of the multigroup procedure for SEM with Jamovi. Do you have anything you can share with me?
I am performing a factor invariance analysis with WLSM, using the SEM multigroup menu. First I estimated the measurement model for one group (model 1). Then I performed the analysis for two unrestricted groups (model 2). Subsequently, I set the constraints of the loadings in the menu (model 3). Jamovi in the output gives me a table with the title "Contraits test scores". It seems to me that it refers to the chi-squared difference obtained for the nested models, I would like to confirm that this is the case, and also if this chi-squared difference is adjusted for WLSM?
I also have some questions about the reference model. For the model with constraints on loadings (model 3), the difference (in the "Contraits test scores" table) is between model 2 (2 groups without constraints) a model 3?. For the model with constraints on the intercepts (model 4), the difference (in the "Contraits test scores" table) is regarding model 2 a model 3.
Many tanks
Tomás
I have been using the SEM module for robust fit indices. Congratulations and many thanks for your work!
I have not found detailed documentation of the multigroup procedure for SEM with Jamovi. Do you have anything you can share with me?
I am performing a factor invariance analysis with WLSM, using the SEM multigroup menu. First I estimated the measurement model for one group (model 1). Then I performed the analysis for two unrestricted groups (model 2). Subsequently, I set the constraints of the loadings in the menu (model 3). Jamovi in the output gives me a table with the title "Contraits test scores". It seems to me that it refers to the chi-squared difference obtained for the nested models, I would like to confirm that this is the case, and also if this chi-squared difference is adjusted for WLSM?
I also have some questions about the reference model. For the model with constraints on loadings (model 3), the difference (in the "Contraits test scores" table) is between model 2 (2 groups without constraints) a model 3?. For the model with constraints on the intercepts (model 4), the difference (in the "Contraits test scores" table) is regarding model 2 a model 3.
Many tanks
Tomás