+ Reply to Thread
Results 1 to 4 of 4

Thread: Confounder

  1. #1
    Omega Contributor
    Points: 38,289, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,992
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Confounder




    What if I see a 10% change (decrease) in an variable effect (X1 on Y) when including another variable (X2), though the effect of X1 is never remotely significant. It is also known X2 is significantly related to Y but not really significantly related to X1 at all.


    The primary outcome is to examine X1 on Y.


    Does this just seem like a confounder issue, any recommendations for things to look at? Secondly, the outcome is binary and rare in this dataset. Currently just modeling Y = X1 + X2. So I will control (condition on) X2.


    With how rare the outcome is I am not too sure how invested I want to get into exploring things, but curious what others think.
    Stop cowardice, ban guns!

  2. #2
    Points: 657, Level: 13
    Level completed: 14%, Points required for next Level: 43

    Posts
    57
    Thanks
    1
    Thanked 9 Times in 8 Posts

    Re: Confounder

    If X2 is found to be a significant predictor of Y, I would certainly include it.

  3. #3
    Omega Contributor
    Points: 38,289, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,992
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Confounder

    Indubitably!
    Stop cowardice, ban guns!

  4. #4
    Points: 3,015, Level: 33
    Level completed: 77%, Points required for next Level: 35

    Posts
    47
    Thanks
    8
    Thanked 7 Times in 7 Posts

    Re: Confounder


    Not sure I know the answer to this, but I am curious to see what others think!

    What is the strength of the association of the variables which each other and with the outcome? Do you actually have quite a strong association but lack power? Can you can increase your sample size- perhaps this is a lot of work!?

+ Reply to 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