+ Reply to Thread
Results 1 to 3 of 3

Thread: Controlling mediation effect in a binary logistic regression

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Location
    Freiburg, Basel
    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Controlling mediation effect in a binary logistic regression




    Hi everbody,

    It's my first post and I hope I posted in the right forum.

    In my master thesis I use binary logistic regression:
    a dichotomous variable as dependent variable, and four ratio scaled variables as independent variables.

    2 of the independent variables turns out as significant predictors (Let's call them Var A and Var B.)

    But there are doubts that there could be a mediation effect. As mediator Var A might affect the relation between Var B and the dependent variable.

    Now to my question. How do I control mediators in a logistic regression? How does a mediation analysis in logistic regression work?

    In a linear regression you can residualize the influence, but I do work with a logistic regression. I paid attention to the correlation between Var A and Var B, which was moderate (0.5). But I am not sure, if that's enough of controlling.

    I'm thankful for every answer.

    Best regards Frodo

  2. #2
    TS Contributor
    Points: 5,246, Level: 46
    Level completed: 48%, Points required for next Level: 104
    maartenbuis's Avatar
    Location
    Konstanz
    Posts
    372
    Thanks
    3
    Thanked 146 Times in 123 Posts

    Re: Controlling mediation effect in a binary logistic regression

    This is actually a surprisingly difficult problem. It is still an area of active research and no concensus has yet been reached. Here is one approach (I am not unbiased): http://www.stata-journal.com/article...article=st0182

  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: Controlling mediation effect in a binary logistic regression


    You may also have some luck looking in genetic research. That is where I have seen some good literature. In particular the presence of certain genes and their mediating effects.
    Stop cowardice, ban guns!

+ 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