+ Reply to Thread
Results 1 to 1 of 1

Thread: Hierarchical regression with partially nested variables

  1. #1
    Points: 5,892, Level: 49
    Level completed: 72%, Points required for next Level: 58

    Location
    Switzerland
    Posts
    27
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Hierarchical regression with partially nested variables




    I would like to do a logistic regression of vaccination coverage of, for example, HPV. Let's assume that we have 50 states (S). 30 states have an urban part and a country part (urbanity variable U). The other 20 have only a country area. 10 of the states have population areas where individuals speaks only Spanish and other areas only English (L). The linguistic region correlates only slightly with urbanity. For example:

    S(1)-U(urban)-L(engl)
    S(1)-U(country)-L(engl)
    S(1)-U(urban)-L(spanish)
    S(1)-U(country)-L(spanish)

    S(2)-U(urban)-L(engl)
    S(2)-U(urban)-L(spanish)

    S(3)-U(urban)-L(engl)
    S(3)-U(country)-L(engl)

    S(4)-U(urban)-L(engl)
    S(4)-U(country)-L(spanish)

    where the number indicate a state.

    The situation is that U is nested in S but not each S has both levels of U. This is also the case for the linguistic variable. If U would be nested completely in S and L completely in S, the model would look like (in R-terminology):
    y ~ (1|S) +(1|S:U) + (1|S:L)
    Is it possible to estimate odds ratio using this model even if they are not completly nested?
    Or Should be used another model, for example:
    y ~ (1|S) + U + L
    y: 0/1 vaccinated
    (1|S) random variable
    U and L fix variable.

    Or does it makes more sense to use only random variables?
    y ~ (1|S) + (1|U) + (1|L)

    Problems which arise using fixed variables: would give significant results or narrow CI due to the high number of individuals.
    I would be very happy if someone could give me some hints how to cope with partially nested variables.
    Last edited by giordano; 07-20-2013 at 08:49 AM. Reason: correction of R-terminology, others

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats