Hello Dr. Geiser, your tutorials are really helpful, thank you!! I have one question concerning the KNOWNCLASS. Mplus seems to treat the known grouping variable as a latent class and therefore adds parameters in the calculation. Do you know why MPLUS is doing so? I also work with R and the calculations when conducting a multigroup LCA always have different parameters, with MPLUS having one parameter more.
Hello, Dr. Geiser. Thank you so much for the informative and thoughtful lectures. I truly appreciate your effort and hardwork putting into Mplus. As I am interested in applying LCA and LTA in my upcoming dissertation, I was wondering about serting the threshold. In your case, those were binary indicators that your syntax was $1 for the variables like action, logic, and etc. What if the indicators were not binary but ordinals? (as some variables are rated as 5-likert scale) In such case, do I write $4 as the threshold?
Glad you like the videos! You would specify as many threshold parameters as there are categories minus one. So for 5 response categories you would have to specify 4 thresholds $1, $2, $3, $4 for each item. Best, Christian Geiser
This video is very interesting, thank you very much! You explain it very well Christian. Now my head is full of ideas and possibilities!! Can we also do this with LPA? Where can we learn to interpret the results? Do you have another video or are you working on it? Thanks =)
Hi Sara, Thank you very much for your positive feedback! Yes, you can also run an LPA as a multigroup analysis. I discuss this in detail in my online LPA workshop which can be found here: www.goquantfish.com/courses/latent-profile-analysis-with-mplus Best, Christian Geiser
Hi! I keep getting the error below, even though my categorical variable is defined in the VARIABLES and USE VARIABLES command. I've even switched variables and whatever variable I put in the categorical command, I keep getting an error. Do you have any idea why? The error is: *** ERROR in VARIABLE command Unknown variable in CATEGORICAL option: immigany
You should send your input and data file to the Mplus support team along with your license number. They will be happy to help you figure this out. Best, Christian Geiser
Hello Dr. Geiser, your tutorials are really helpful, thank you!! I have one question concerning the KNOWNCLASS. Mplus seems to treat the known grouping variable as a latent class and therefore adds parameters in the calculation. Do you know why MPLUS is doing so? I also work with R and the calculations when conducting a multigroup LCA always have different parameters, with MPLUS having one parameter more.
Hello, Dr. Geiser. Thank you so much for the informative and thoughtful lectures. I truly appreciate your effort and hardwork putting into Mplus. As I am interested in applying LCA and LTA in my upcoming dissertation, I was wondering about serting the threshold. In your case, those were binary indicators that your syntax was $1 for the variables like action, logic, and etc. What if the indicators were not binary but ordinals? (as some variables are rated as 5-likert scale) In such case, do I write $4 as the threshold?
Glad you like the videos! You would specify as many threshold parameters as there are categories minus one. So for 5 response categories you would have to specify 4 thresholds $1, $2, $3, $4 for each item.
Best, Christian Geiser
This video is very interesting, thank you very much! You explain it very well Christian. Now my head is full of ideas and possibilities!!
Can we also do this with LPA? Where can we learn to interpret the results? Do you have another video or are you working on it? Thanks =)
Hi Sara,
Thank you very much for your positive feedback! Yes, you can also run an LPA as a multigroup analysis. I discuss this in detail in my online LPA workshop which can be found here:
www.goquantfish.com/courses/latent-profile-analysis-with-mplus
Best, Christian Geiser
Hi! I keep getting the error below, even though my categorical variable is defined in the VARIABLES and USE VARIABLES command. I've even switched variables and whatever variable I put in the categorical command, I keep getting an error. Do you have any idea why? The error is:
*** ERROR in VARIABLE command
Unknown variable in CATEGORICAL option:
immigany
You should send your input and data file to the Mplus support team along with your license number. They will be happy to help you figure this out.
Best, Christian Geiser