+ Reply to Thread
Results 1 to 4 of 4

Thread: Large odds ratio in binary logistic regression/big scale difference of variables

  1. #1
    Points: 450, Level: 8
    Level completed: 50%, Points required for next Level: 50

    Posts
    8
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Question Large odds ratio in binary logistic regression/big scale difference of variables




    I'd like to ask for some help with a binary logistic regression. In SPSS I am building a binary logistic regression with 4 independent continuous variables(Sample size - 85). However, with one of the variables (Bicaudatus_index) I get a huge odds radio:


    I think that I know where the problem is: the scale of this variable is very different than other variables:


    As this variable is a ratio of two measurements, I try to multiply the variable 100 times and get a new variable, which odds ratio in the same regression seem to be within normal range.


    However I don't know if it is appropriate to do that. If I multiply the variable not 100, but e.g. 150 times, I get odds ratios that are different from the ones that I get with 100.

    Therefore, I want to ask what should I do in this case, is this method appropriate or do I need to interpret the results differently because of multiplication, or do I need a whole new approach. Help would be greatly appreciated

  2. #2
    TS Contributor
    Points: 12,227, Level: 72
    Level completed: 45%, Points required for next Level: 223
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,470
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Large odds ratio in binary logistic regression/big scale difference of variables

    hi,
    this is quite natural and you are on the good track. The odds ratio tells us how the odds change if we change the X by one unit. Imagine we look at odds of heart attack vs. weight -the odds will definitely be greater if we measure the weight in 50 kg units as when we measure in grams (i.e the odds of a heart attack by a weight increase of 50kg vs. a weight increase of 1 g)
    Regards

  3. The Following User Says Thank You to rogojel For This Useful Post:

    gianmarco (10-02-2016)

  4. #3
    Points: 450, Level: 8
    Level completed: 50%, Points required for next Level: 50

    Posts
    8
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Re: Large odds ratio in binary logistic regression/big scale difference of variables

    Quote Originally Posted by rogojel View Post
    hi,
    this is quite natural and you are on the good track. The odds ratio tells us how the odds change if we change the X by one unit. Imagine we look at odds of heart attack vs. weight -the odds will definitely be greater if we measure the weight in 50 kg units as when we measure in grams (i.e the odds of a heart attack by a weight increase of 50kg vs. a weight increase of 1 g)
    Regards
    Thank you for your answer . So as I understand, it is appropriate to multiply the varible e.i. 100 times to get a valid odds ratio, this way one just has to interpret it differently?

  5. #4
    TS Contributor
    Points: 12,227, Level: 72
    Level completed: 45%, Points required for next Level: 223
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,470
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Large odds ratio in binary logistic regression/big scale difference of variables


    I would phrase it differently - it is ok to re-scale the variable so that the change of 1 unit is of practical interest. E.g. neither a weight gain of 1 gram nor of 50kg is practically interesting, but maybe the effect of a change ofb500g or 1 kg would be.

    regards

  6. The Following User Says Thank You to rogojel For This Useful Post:

    gianmarco (10-02-2016)

+ 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