+ Reply to Thread
Results 1 to 5 of 5

Thread: Priori power calculation

  1. #1
    Points: 707, Level: 14
    Level completed: 7%, Points required for next Level: 93

    Posts
    6
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Priori power calculation




    I'm about to conduct a study as part of my PhD and am a bit confused on my supervisors advice on conducting a priori power calculation to determine sample size.

    Essentially we know we're going to be using 16 participants, so we are using data from previous research to work out what size of effect we detect with a size of 16. My supervisors advice was this:

    "Change this to reflect the higher sample size ie reverse engineer it so you work out what size of effect you can detect with a sample size of 16. Report a range of effect sizes if possible"

    Can anyone guide me on how I go about doing this using G*power?

  2. #2
    TS Contributor
    Points: 12,287, Level: 72
    Level completed: 60%, Points required for next Level: 163
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,471
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Priori power calculation

    hi,
    in power calculations you have 3 variables (power, effect, sample size) and by fixing 2 you can calculate the third. Of course you always have to have an idea of the population standard deviation.

    So, what is asked from you is to use the power calculation program and input the sample and the power (80% I presume) to see what is the smallest effect size you can hope to detect (provided you estimate the standard deviation well enough)

    regards

  3. #3
    Points: 707, Level: 14
    Level completed: 7%, Points required for next Level: 93

    Posts
    6
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Priori power calculation

    Quote Originally Posted by rogojel View Post
    hi,
    in power calculations you have 3 variables (power, effect, sample size) and by fixing 2 you can calculate the third. Of course you always have to have an idea of the population standard deviation.

    So, what is asked from you is to use the power calculation program and input the sample and the power (80% I presume) to see what is the smallest effect size you can hope to detect (provided you estimate the standard deviation well enough)

    regards
    Thanks very much for your help. I think I'm a little closer to getting it right but I'm still a little confused. If I fix two of the variables (power and sample size), then how does the effect sizes I have gained from previous research that is similar in nature come into it? I have effect sizes for the variables of interest based on previous research, so I'm wondering how these have any influence in my calculation?

  4. #4
    Points: 707, Level: 14
    Level completed: 7%, Points required for next Level: 93

    Posts
    6
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Priori power calculation

    bump...........

  5. #5
    TS Contributor
    Points: 12,287, Level: 72
    Level completed: 60%, Points required for next Level: 163
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,471
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Priori power calculation


    They don't have any effect, but you can compare what you get from the calculation to what previous experiments found. E.g. if your minimal effect size based on the power calc is way larger then what has already been found, you might conclude that you need more samples ...and so on.

    regards

+ 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