+ Reply to Thread
Results 1 to 4 of 4

Thread: It's posible to compare two means without considering standard deviations?

  1. #1
    Points: 90, Level: 1
    Level completed: 80%, Points required for next Level: 10

    Posts
    9
    Thanks
    0
    Thanked 1 Time in 1 Post

    It's posible to compare two means without considering standard deviations?




    My doubt is the following

    'it's possible to compare means of two independent groups without considering standard deviations" ?
    For example:

    Control Group: mean +- standard deviation 50 +- 10 mm
    Experimental Group: mean +- standard deviation 100 +- 30 mm

    I am very surprised with the following conclusion made by a dentist friend mine

    "Then we say there was an increase of mean 100% between groups !!"

    (100 - 50) / 50 = 1 = 100%

    However this conclusion seems odd to me, as it disregardes the variability of the data.
    TIA
    Ivan

  2. #2
    Omega Contributor
    Points: 39,173, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,088
    Thanks
    405
    Thanked 1,196 Times in 1,157 Posts

    Re: It's posible to compare two means without considering standard deviations?

    Is treatment randomized?


    Were groups the same size?


    What does the continuous variable represent?


    This seems wonky:


    "(100 - 50) / 50 = 1 = 100%"


    100 / 50 = 2 times or 200%
    Stop cowardice, ban guns!

  3. #3
    Points: 90, Level: 1
    Level completed: 80%, Points required for next Level: 10

    Posts
    9
    Thanks
    0
    Thanked 1 Time in 1 Post

    Re: It's posible to compare two means without considering standard deviations?

    Yes. The treatment can be randomized. The same site sample, for example. The variable can tô be mm, for example.


    Quote Originally Posted by hlsmith View Post
    Is treatment randomized


    Were groups the same?



    What does the continuous variable represent?


    This seems wonky:


    "(100 - 50) / 50 = 1 = 100%"


    100 / 50 = 2 times or 200%

  4. #4
    Points: 90, Level: 1
    Level completed: 80%, Points required for next Level: 10

    Posts
    9
    Thanks
    0
    Thanked 1 Time in 1 Post

    Re: It's posible to compare two means without considering standard deviations?


    The groups are differents
    For example, ten subjects for Control and others ten subjects for Experimental.
    .....................................
    Quote Originally Posted by balducci View Post
    Yes. The treatment can be randomized. The same site sample, for example. The variable can tô be mm, for example.

+ 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