+ Reply to Thread
Results 1 to 4 of 4

Thread: sample size

  1. #1
    Points: 17, Level: 1
    Level completed: 33%, Points required for next Level: 33

    Posts
    2
    Thanks
    1
    Thanked 0 Times in 0 Posts

    sample size




    Hi everyone,

    I do my research to see the association of 2 qualitative variables which make 2x3 contingency table. I calculate the required sample size by using G*Power. I got 964 of the total sample size from G*Power. The problem is my population size have just 330. What should I do?

    For more information, the parameter that I input in G*Power is as followed:

    - effect size = 0.1
    - alpha=0.05
    - power=0.8
    - df=2.

    Thank you.

  2. #2
    Omega Contributor
    Points: 39,022, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,069
    Thanks
    402
    Thanked 1,192 Times in 1,153 Posts

    Re: sample size

    Given that effect size, what power will you have? Now is that risk of a type II error (accepting the null when it should be rejected) acceptable to you?


    You could also see what effect size given your constraints that you can reliably discern giving the constraints.
    Stop cowardice, ban guns!

  3. #3
    Points: 17, Level: 1
    Level completed: 33%, Points required for next Level: 33

    Posts
    2
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Re: sample size

    Quote Originally Posted by hlsmith View Post
    Given that effect size, what power will you have? Now is that risk of a type II error (accepting the null when it should be rejected) acceptable to you?


    You could also see what effect size given your constraints that you can reliably discern giving the constraints.

    The effect size of 0.1 that I set was from the previous research. Sometimes I set it from my pilot research. Can it from other way? Can I set it by myself?

  4. #4
    TS Contributor
    Points: 17,949, Level: 85
    Level completed: 20%, Points required for next Level: 401
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,561
    Thanks
    56
    Thanked 643 Times in 605 Posts

    Re: sample size


    If previous research indicates an only small effect size (0.1), then using a larger effect size for the calculations would look dubious, IMHO.
    What you could possiby do is to calculate the power you'd achieve with n=330, effect size 0.1 and alpha = 0.05; so that you can get an
    impression whether power is still acceptable.

    With kind regards

    PonderStibbons
    »Jetzt kann mich der Führer mal am Arsch lecken.« (Ernst Kuzorra, 1941)

  5. The Following User Says Thank You to Karabiner For This Useful Post:

    saitharndham (11-06-2017)

+ 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