What is the required sample size for latent class cluster analysis for 9-12 indicators?

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!

Hi, i am using multiple regression for one of my papers and i came across two types of B; B  and β, for which i don't understand the difference. Looking for answers in books and internet made it worse. Some say that β is the power, or similar to the alpha level, and some say that it is the beta weight. But if β is the beta weight, then what does the B stands for?

Thanks in advance

I am conducting Latent Class Analysis with 8 binary indicators and a sample size of 6757 obs. The entropy for 2classes, 3classes, 4 classes clustering is very low (around 0.54). But when I reduce the sample size and drop some of the observations randomly, I obtain better entropies. For example the entropy for 4500 sample size is 0.68 and for 4280 sample size is 0.75.( For all of the sample sizes BIC suggests that 3 classes clustering is the best number of clustering). I am wondering is it acceptable to reduce the sample size to get a better entropy?! Is there any justifications to reduce sample size in this case ? Any hint would be of great help. Thanks

Hi everybody,

I have a moderator variable, since I have 2 predictors and 4 dependent variables, I was thinking of SEM. But my supervisor says that I can't have a moderator variable using SEM. Is that true?

Thanks in advance for your time and help

I need to know is there is a minimun number of subject to run a cluster analysis. My sample has 546 subjects, y I have two test of five factors each one. A revierw ask me why I think that this number is enough, it would be better if I can do it quoting some research.

Dear colleagues,

I conducted my LCA in Mplus in order to find out how many types of prospective teachers I can find with regard to their self-efficacy beliefs for teaching the German subject "Gesellschaftslehre" (Civic Education, History and Geogaphy combined). My total n = 112.

The best result according to BIC, adj.-BIC, Hitrate and BLRT is 4 classes.

1. class (n=23)

2. class (n=35)

3. class (n=47)

As you can see, the final class consists only of four people. Yet, content-wise it is a very interesting one.

My question: Would you consider a class size of 4 too small regardless or would you consider is permissable because of what characterises this particular group?

Literature references and/or your opinions are, as always, highly welcome.

Thank you in advance!

Marcel

I have heard on multiple occasions that three indicators in a LCA will likely result in a high, medium, and low group. Therefore if you have 2 or 3 indicators, a cluster analysis is preferred. However, I cannot find a reference for this anywhere. The only reference I have been able to locate on number of indicators in a LCA has a minimum of 4. Does anyone have a good reference for this point?

Thanks!

I wish to know how to test reliability if the questionnaire includes both dichotomous and 5 point likert scale items. is cronbach alpha helpful for both? what to do when alpha score is less than 0.7?