+ Reply to Thread
Results 1 to 4 of 4

Thread: BMI as grouped/continous variable?

  1. #1
    Points: 150, Level: 2
    Level completed: 50%, Points required for next Level: 50

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    BMI as grouped/continous variable?




    I have a question the BMI variable:

    What are the pros and cons for using BMI as a variable grouped after WHO's classification of BMI (>18.5, 18.5-25, 25-30, <30) instead of using it as a continous variable?

    Thank you.

  2. #2
    Points: 1,838, Level: 25
    Level completed: 38%, Points required for next Level: 62

    Location
    UK
    Posts
    163
    Thanks
    0
    Thanked 9 Times in 9 Posts

    Re: BMI as grouped/continous variable?

    A good thing is that those groupings well understood meanings so they can be interpreted. A disadvantage is that, as with all categorisation really, you are throwing away data you have.

    Really, its a decision you have to make depending on the analysis you are doing.

  3. #3
    Points: 1,838, Level: 25
    Level completed: 38%, Points required for next Level: 62

    Location
    UK
    Posts
    163
    Thanks
    0
    Thanked 9 Times in 9 Posts

    Re: BMI as grouped/continous variable?

    This paper about categorisation that I saw recently is quite good: http://www.biomedcentral.com/content...2288-12-21.pdf

  4. #4
    RoboStataRaptor
    Points: 10,079, Level: 67
    Level completed: 8%, Points required for next Level: 371
    bukharin's Avatar
    Location
    Sydney, Australia
    Posts
    1,284
    Thanks
    11
    Thanked 317 Times in 308 Posts

    Re: BMI as grouped/continous variable?


    Here's a slightly different take - why use BMI at all?
    Kronmal RA. Spurious correlation and the fallacy of the ratio standard revisited. Journal of the Royal Statistical Society. Series A (Statistics in Society). JSTOR; 1993;:379–92.

    Abstract:
    Spurious correlation refers to the correlation between indices that have a common component. A 'per ratio' standard is based on a biological measurement adjusted for some physical measurement by division. Renowned statisticians and biologists (Pearson, Neyman and Tanner) have warned about the problems in interpretation that ratios cause. This warning has been largely ignored. The consequences of using a single ratio as either the dependent or one of the independent variables in a multiple-regression analysis are described. It is shown that the use of ratios in regression analyses can lead to incorrect or misleading inferences. A recommendation is made that the use of ratios in regression analyses be avoided.

    If you do decide to use BMI as a continuous variable you'll need to check your regression model carefully for violations of assumptions - BMI often has a (very) nonlinear relationship with whatever else you're measuring.

+ 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