Page 1 of 1

error messages with LCA and LPA with SnowRMM

Posted: Sat Oct 14, 2023 12:07 pm
by RMD
Hello people,
I've been getting an error message when trying to perform a latent class analysis and a latent profile analysis, which I don't understand. The data is categorical.

For the LPA, the error message is: "Estimate profiles can only handle numeric values. The variables violating this assumption are..." And the variables listed a nominal. but integers.

For the LCA, the error message is: "dim(x) must have a positive length" (using the same variables).

Is anyone able to help with this? (The data were imported from Excel, and other analyses work).

Many thanks!
Russell

Re: error messages with LCA and LPA with SnowRMM

Posted: Sun Oct 15, 2023 12:45 am
by seol
Hi

Could you attach a file that fails analysis here?

Cheers
Seol

Re: error messages with LCA and LPA with SnowRMM

Posted: Sun Oct 15, 2023 2:50 am
by RMD
Hi Seol,
Sure, here's the file:
https://1drv.ms/u/s!Ak18NDjxZnZr6kcHffa ... 9?e=fE6SEw
Many thanks,
Russell

Re: error messages with LCA and LPA with SnowRMM

Posted: Sun Oct 15, 2023 6:46 am
by seol
Hi Russell

First, for the Latent Class analysis, the variables should be coded from 1 to maximum number, not from 0.
Second, for the Latent Profile analysis, the measure type of variables that is analyzing is continuous, not nominal.

I attached a file with the error fixed here.

Best Regards
Seol

Re: error messages with LCA and LPA with SnowRMM

Posted: Sun Oct 15, 2023 8:53 am
by RMD
Many thanks Soel!
That's very kind of you to fix it as well as explain where I went wrong.
Best regards,
Russell