intraclass correlation

Home Forums Methodspace discussion intraclass correlation

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #1777
    Kay Smith
    Member

    Another power calculation of sorts.

    I have calculated an ICC 0f 0.886

     

    When trying to calculate the power of this study I need to know

    R0 (null hypotheis) and R1 (perfect hypothesis)

    Is the null hypothesis that there is NO correlation and therefore R0=0 and my alternative hypothesis is 0.886 that i calculated above?

     

    There are a few worked examples on the web but they have R0=0.2

    How would you be able to estimate a precise fugure for a null hypothesis?

    #1778

    I am not an expert in ICCs, but having 0 as the null hypothesis is simply a convention (and not a good one, as many have pointed out before). If theory tells you that the null hypothesis should include something else than 0, it is something else than 0. This means the example you found on the web should work equally well with your data.

Viewing 2 posts - 1 through 2 (of 2 total)
  • You must be logged in to reply to this topic.