+ Reply to Thread
Results 1 to 5 of 5

Thread: Odds ratios lower than 1 for continous variable

  1. #1
    Points: 2,465, Level: 30
    Level completed: 10%, Points required for next Level: 135

    Posts
    12
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Odds ratios lower than 1 for continous variable




    Lets says you want to check the effect of age on a specific event.

    In the univariate model : event = age, the odds is significantly lower than 1 which seems to show a 'protective impact' of age on this event (age increase -> probablity of event decrease)

    However, after checking, average of age is significantly higher in peole who had the event than in people who didn't have the event.

    How is it possible ??

    Thanks for your help

  2. #2
    Cookie Scientist
    Points: 13,431, Level: 75
    Level completed: 46%, Points required for next Level: 219
    Jake's Avatar
    Location
    Austin, TX
    Posts
    1,293
    Thanks
    66
    Thanked 584 Times in 438 Posts

    Re: Odds ratios lower than 1 for continous variable

    Make sure your model is modeling the probability of the event occurring (event=1) rather than not occurring (event=0). Some programs such as SAS do the latter by default, which will reverse the signs of the odds ratios.
    “In God we trust. All others must bring data.”
    ~W. Edwards Deming

  3. #3
    Omega Contributor
    Points: 38,396, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,001
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Odds ratios lower than 1 for continous variable

    What is your reference value in the OR, or are you using a bta coefficient from the logistic model?
    Stop cowardice, ban guns!

  4. #4
    Points: 2,465, Level: 30
    Level completed: 10%, Points required for next Level: 135

    Posts
    12
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Re: Odds ratios lower than 1 for continous variable

    Thansk you for your help!
    In the logisitic model of SAS (proc logistic), you have to add DESCENDING in aim to model the propability to have the value 1 if the event is coded 0/1

  5. #5
    Omega Contributor
    Points: 38,396, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,001
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Odds ratios lower than 1 for continous variable


    Or

    (Event = "1")
    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