+ Reply to Thread
Results 1 to 2 of 2

Thread: Power analysis with one and two variables?

  1. #1
    Points: 8, Level: 1
    Level completed: 15%, Points required for next Level: 42

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Power analysis with one and two variables?




    Hi gang,

    I am severely illiterate in stats and so am struggling with what would seem to be a very straightforward problem.

    My problem: We have an autologous tissue construct and need to know what sample size (number of patients) is required to achieve statistical significance (p≤0.05) given a single variable which is "did the tissue graft successfully take/survive post-surgical placement?" (I believe our null would be that the tissue graft would not take in a statistically significant manner - correct?)

    A follow up on the original problem, this time with two variables. What sample size would be required (number of patients) to achieve statistical significance (p≤0.05) given the dichotomous variables (I hope I'm using that phrase correctly) which are "did the tissue graft successfully take/survive post-surgical placement AND did that successful graft display hair follicles?"

    I know what we want our p-value to be but I don't know how to determine confidence intervals which, if I am understanding this correctly, are required to determine the sample size required to achieve (p≤0.05).

    I hope these questions were clear enough given my lack of ability to explain. If you have questions, I'll do my best to answer them!

    Thank you for any light you might be able to shed my way. I'm grateful in advance to any help...

    AB

  2. #2
    Points: 981, Level: 16
    Level completed: 81%, Points required for next Level: 19

    Location
    New Zealand
    Posts
    128
    Thanks
    2
    Thanked 26 Times in 25 Posts

    Re: Power analysis with one and two variables?


    Just getting the simple case worked out first, you will eventually end up with a percentage grafts that worked, and a sample size. What are you planning to test this sample percentage against to get a p value? kat

+ 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