My population is all people above 18 years old in the country, so I think this is an infinite population. Is it suitable to use Cochran formula for sample size calculation while I want to use SEM in analysis?
Observed items as well as parameters are usually used in sample size calculation in SEM studies. According to Hair et al. not only parameters or observed items but also model complexity and number of constructs with communalities should be focused for sample size calculation. They recommended 15 respondents for each parameter in the model to maintain multivariate normality. They also mentioned that minimum sample size should be at least 500 if the model with large numbers of constructs (>7) and some with lower communalities and/or having fewer than three measured items.
Hair JF, Black WC, Babin BJ, Anderson RE. Structural equations modeling overview. Multivariate Data Analysis, 7th edition. United Kingdom: Prentice Hall PTR; 2013.
Please refer to the attached research gate link. You might find the interactions therein useful. https://www.researchgate.net/post/Is_that_possible_to_use_Cochran_formula_sample_size_determination_for_heterogeneous_groups_populations
There's no consensus about this. However, a 384 sample size would be acceptable regardless of the population size—also, the bigger, the better. Besides, the choice of the sampling method would be critical regarding the representation of the population by the sample. See more details in this article.
Article Developing and Validating a New Multi-Dimensional Scale for ...
I agree with most of the answers especially from Muhammed Ashraful Alam . OIne estimate is to have 10-20 respondents per parameter is considered satisfactory to employ SEM in your study.