LCA - running the same analysis with different results

General help and assistance with jamovi. Bug reports can be made at our issues page: https://github.com/jamovi/jamovi/issues . (If you're unsure feel free to discuss it here)
Post Reply
02ysoe
Posts: 3
Joined: Mon Aug 15, 2022 6:48 pm

LCA - running the same analysis with different results

Post by 02ysoe »

Hi there,

I'm new to LCA and Jamovi, so hoping for a little help.

Running the same analysis with the same dataset, but trying out different number of class setting, gives me different results every time. Not only class probabilities, but also model comparison output. Consider this specific example:

- First running analysis with 4 classes which gives me attached output 1.

- Then running the analysis with 3 classes which gives me attached output 2.

- Then back to running with 4 classes with output 3.

- And finally running with 3 classes again with output 4.

EDIT: Whooops - can only attached 3 files...

Why do they differ (quiet a lot) when running the exact same analysis and just changing the number of classes I'm looking for? Again, I'm new to it, and the may be an obvious explanation that I'm missing.

Thx
Attachments
Output 1.PNG
Output 1.PNG (7.51 KiB) Viewed 677 times
Output 2.PNG
Output 2.PNG (6.52 KiB) Viewed 677 times
Output 3.PNG
Output 3.PNG (7.74 KiB) Viewed 677 times
User avatar
jonathon
Posts: 2613
Joined: Fri Jan 27, 2017 10:04 am

Re: LCA - running the same analysis with different results

Post by jonathon »

if you get different results each time, it typically means the analyses are stochastic, and that's to be expected. but it shouldn't dramatically change the interpretation.

cheers
Post Reply