'Matching with multi-level multiple membership data

I am designing a within multi-level study in which my data has both nested and multiple-membership structure. The subjects are multimembers of V1 and also are nested in V2.The subjects are all from one year. When applying matching or weighting, should I take the multiple membership into account? if yes, the question would be how? or should I apply matching without taking it into account and then run a multiple membership multi-level regression on the output from matching?



Solution 1:[1]

Without knowing more about your concrete use case, it's hard to recommend specific sources. But check out "cross-classified multilevel analysis". I think that's what you're referring to.

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source
Solution 1 guest