+ Reply to Thread
Results 1 to 3 of 3

Thread: Can a post-hoc power analysis be done?

  1. #1
    Points: 540, Level: 10
    Level completed: 80%, Points required for next Level: 10

    Posts
    21
    Thanks
    6
    Thanked 0 Times in 0 Posts

    Can a post-hoc power analysis be done?




    Hi Guys,
    Just wondering if anyone could help please. For the following data I don't think a post-hoc power analysis can be done?

    In 2014 the number of referred patients who had the event was 22/55. Whilst in 2008 it was 15/55 who had the event.


    2014: 22 events vs 33 no events
    2008: 15 events vs 40 no events

    2 independent samples

    Events in 2008 n=15, Events in 2014 n=22
    Chi squared 1.99. df=1, p=0.15

    Because no statistical significance was found ie p=0.15 I contacted a post-hoc power analysis based on a 95% confidence interval and 5% error margin and found it to be 0.65? Is this correct?

    And should I have performed a sample size calculation prior to conducting the 2014 study?

    Many thanks for your help

  2. #2
    TS Contributor
    Points: 12,227, Level: 72
    Level completed: 45%, Points required for next Level: 223
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,470
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Can a post-hoc power analysis be done?

    hi,
    I think the point of the post hoc power analysis is to calculate the smallest effect size that would have been detectable with your sample. You cam use this as an upper limit of the effect if your test was not significant. What was the research question in your case and the effect size?

    regards

  3. #3
    Points: 540, Level: 10
    Level completed: 80%, Points required for next Level: 10

    Posts
    21
    Thanks
    6
    Thanked 0 Times in 0 Posts

    Re: Can a post-hoc power analysis be done?


    Hi There
    The observed effect was 27% (15/25=events) versus 40% (22/55 events)
    But a difference of interest, like 20% (say 27% versus 47%) would have been better.
    But not sure how how much power this would have yielded for that with 55 subjects per group.

+ 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