Good day! I have a sample size of 76 youth and have been advised that LPA would not be good to use with this sample size in MPlus. Is there a recommended sample size for latent profile analyses and why? I have some theories but I really want to understand. Also, can anyone recommend any good article that discusses sample size, statistical power and LPA? Thank you!
Good question! Please find the papers as attached.
http://www.ncbi.nlm.nih.gov/pubmed/24489457
http://www.tqmp.org/Content/vol03-2/p043/p043.pdf?origin=publication_detail
Hello Judith,
I suppose similarly as for SEM analysis, 200 participants is usually suggested as minimum and 300 is closer to optimal sample size.
Best wishes
Khandoker, thank you for the articles which I am excited to read. Witold, you know interestingly when I was at a statistical training, I learned that the minimum for SEM was 120 participants and that after 120, having more participants does not significantly increase your ability to determine effect size. Have you heard of this?
There are many approached and opinions by different authors, but if you plan to use pls-sem, then the minimum sample size could even be lower. Of course, that does not mean the minimum sample size is recommended, that is just a general guideline.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Applied Statistics