I'm interested in determining if there is an association between vitamin a levels (continuous) and stunting (nominal/ ordinal), in children (age 6-59 months, who will be split into smaller sub-age groups e.g 6-12, 12-24 etc), sample size ~ 1000. The data is hierarchical because of stratified cluster sampling. In the sampled households, all preschool children had their data collected, meaning there are some sibling pairs in some households and single samples in others. I would also want to factor in the effect of household-level vitamin A fortificants (continous/ordinal) on the vitamin A levels, in my analysis.

I've been reading around mixed models and find myself more confused. I am not sure if this would be the right approach (since I may not be able to prove a linear relationship between the two variables) as opposed to the traditional correlation /chi-square tests.

Any suggestions (even to less confusing reading material) will be appreciated.