+ Reply to Thread
Results 1 to 4 of 4

Thread: Fisher test or interaction terms to determine if 2 groups are different

  1. #1
    Points: 220, Level: 4
    Level completed: 40%, Points required for next Level: 30

    Posts
    4
    Thanks
    1
    Thanked 1 Time in 1 Post

    Post Fisher test or interaction terms to determine if 2 groups are different




    Hi,

    I split my sample in 2 groups, participants with condition +M and participants without the condition -M. I calculated correlations and beta coeff( regression by group) so what is the better way to see if these 2 groups differ significantly? the Fischer test for correlations or interaction terms in the regression? Please let me know if I need to provide more details?

    Thanks in advance for your help.
    Iris

    PS I am new to this site, how can I give back seeing as I am no expert on stats?

  2. #2
    Omega Contributor
    Points: 38,303, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,993
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Fisher test or interaction terms to determine if 2 groups are different

    Provide more information!!

    What is your dependent variable and how is it formatted?

    Fisher's for correlation - not sure that exists?

    If you are running a regression, what is this third variables you have not mentioned?
    Stop cowardice, ban guns!

  3. #3
    Points: 220, Level: 4
    Level completed: 40%, Points required for next Level: 30

    Posts
    4
    Thanks
    1
    Thanked 1 Time in 1 Post

    Re: Fisher test or interaction terms to determine if 2 groups are different

    Hi,

    Thanks for wanting to know more. So I am looking at the association/correlation of cortisol (independent) to white blood cells (dependent), and my outcome is the condition M+. So if I look at correlation between cortisol and WBC, it is sig in M- group and non sig in M+ group. But I want to check if it differs between the 2 groups and I thought I would look at Fisher r to Z transformation (I used this http://vassarstats.net/rdiff.html) I did this and indeed there is a sig diff. I also looked at the regression where I coded it in STATA as: regression WBC cortisol M cortisol*M (covariates) and the interaction term cortisol*M was NOT significant.

    Thanks for all your help!

  4. #4
    Omega Contributor
    Points: 38,303, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,993
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: Fisher test or interaction terms to determine if 2 groups are different


    So to paraphrase, you have two independent variables both continuous (cortisol and WBC) in a model predicting a binary condition. You want to know if they interact (multiplicative term). I would just keep them in the model plus their multiplicative interaction term. I would also plot Cortisol by WBC with the sample stratified by binary outcome, and the two lines should not be parallel and could visibly cross.
    Stop cowardice, ban guns!

+ 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