Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Three way interaction or multilevel

    Hey everyone!

    I was just wondering if any of you have any advice to give regarding the advantage of different models. My regression analysis is based on a dataset of 20 countries examining two-way interactions of individual-level variables (which are significant) but I would like to examine how these interactions depends on the region. I have divided the sample into three regions (7 countries in each) and in separate analyses for each region tested the interaction effects. The interactions do seem to differ between the regions. However, would it be better to run it in the same analysis and make a three-way interaction where the third variable is the region? I suppose I can't tell if there are significant differences between the regions when I'm solely looking at 7 countries at a time?

    /Johan

  • #2
    Johan:
    three-way interactions are oftentimes difficult to expalin/disseminate.
    It's probably better in your case going -mixed- (individuals nested within regions which, in turn, are nested within countries).
    Kind regards,
    Carlo
    (Stata 16.0 SE)

    Comment


    • #3
      Thank you so much Carlo for sharing your thoughts! So it is possible to see if the interaction effects vary in different regions through a multilevel model?

      Comment


      • #4
        Johan:

        that's the idea if you have a nested study design.
        Kind regards,
        Carlo
        (Stata 16.0 SE)

        Comment

        Working...
        X