[Paras Mehta]

Your data and model correspond to the example on the help page:
http://xxm.times.uh.edu/learn-xxm/latent-growth-curve-model/

Please take a look at the data for the example. Here is the key change you would need to make:
The first column must be IDs for the specific level (‘response’). In your case, the first column is a predictor. Create a new column called ‘response’. Each row of ‘response’ will be a unique integer value. ‘response’ will also be the the name of the first level. Doing so allows xxM to link data from the ‘response’ level to student level data.

hope this helps.

[Paras Mehta]

Yes, it is possible. The matrix to use would be a beta matrix. If a latent factor V is defined by other latent variables U_1 to U_p, a (p by 1) beta matrix with one of the paths fixed to 1.0 will do the trick.

Simplest case is the ‘hierarchical random-intercept’:

Three-Level Hierarchical Model with Across-Level Latent Variable Regression

Here is an example of random-slopes for a latent DV.

Two-Level Confirmatory Factor Analysis with a Random Slope

In each case, a factor is defined by other latent factors.

Best,

Paras

[Paras Mehta]

Hi,

1. xxM 0.6.0 is compatible with R 3.2.0. You can install the package using:

install.packages("C://path-to-download//xxm.zip", repos=NULL).

Alternatively, you could use Rstudio to install the package.

2. A new version of xxM compiled for various versions of R will be released shortly. It will include new features.

Best,

Paras

[Paras Mehta]

For now, citation in APA format should be
Mehta, P. D. (2013). xxM User’s Guide. Retrieved from http://xxm.times.uh.edu/
Mehta, P. D. (2013). xxM [Computer software]. Retrieved from http://xxm.times.uh.edu/

[Paras Mehta]

Hi Bill,

Thanks for the post. Could you please email me your script?

paras [dot] mehta [at] times [dot] uh [dot] edu

Paras

[Paras Mehta]

Hi,

Current version allows only continuous MVN outcome.
Eventually! 🙂

Paras

[Paras Mehta]

Thank you all,

I will put Mac on the list.

Paras

[Paras Mehta]

Thanks for the feedback and sorry for the confusion.

Neither the model object (brim) nor the dataframe object (brim.student) should be in quotes.

Could you please try to run the script included with the package? You can find the script under “\xxm\models\brim” directory.  Once the packaged data is loaded using data() command:

data(brim.xxm, package = "xxm")

you can examine the contents by issuing the str() command:

str(brim.student)
str(brim.teacher)

Please let me know if you have any trouble with script included with the package. The rest of the script should work if both datasets are loaded.

 

• This reply was modified 10 years, 10 months ago by Paras Mehta.
• This reply was modified 10 years, 10 months ago by Paras Mehta.
• This reply was modified 10 years, 10 months ago by Paras Mehta.
• This reply was modified 10 years, 10 months ago by Paras Mehta.
• This reply was modified 10 years, 10 months ago by Paras Mehta.

[Paras Mehta]

Hi Mike,

Source package is not available for download at this time. I will post a binary package shortly.

Paras

[Paras Mehta]

Thanks Mark,

Made the correction.

Paras

[Author]

Posts