+ Reply to Thread
Results 1 to 2 of 2

Thread: general mediation question

  1. #1
    Points: 25, Level: 1
    Level completed: 49%, Points required for next Level: 25

    Thanked 0 Times in 0 Posts

    general mediation question

    Hi guys

    Probably the easiest question ever but I can't really wrap my head around it. The setup of the research is as follow:

    X and M = continuous
    Y = dichotomous

    Suppose you are testing mediation and you have:

    X --> Y insignificant (logistic regression)

    X --> M significant (using non-parametric test because M failed all residual normality assumptions and M is not dichotomous; so I transformed X and M into categories and use chi-square test just to look at association)

    M --> Y significant (logistic regression)

    Would this be something like full mediation, partial or no at all? And is it possible what I have done, like the setup of two different tests? Would you be able to conclude something from this and if so what?

    Hope you can help me, I just need some general answers which is why I did not go in depth on the description of variables and just want to know what you think/would do in this setting. I am using SPSS. Thanks a bunch!

  2. #2
    Omega Contributor
    Points: 38,303, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Not Ames, IA
    Thanked 1,185 Times in 1,146 Posts

    Re: general mediation question

    Could you normalize with log or squaring? Not too experienced in mediation, but it seemed like you typically have to calculate ide, cde, and a few others, you may have down that but without the lingo.

    If I follow x predicts m. M predicts y and x does not predict y. What type of study design do you have?
    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