+ Reply to Thread
Results 1 to 6 of 6

Thread: Identifying Effect size in Power Analysis

  1. #1
    Points: 3,628, Level: 37
    Level completed: 86%, Points required for next Level: 22

    Posts
    11
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Identifying Effect size in Power Analysis




    I have a set of data for a water quality parameter on which is used to compute Rank Spearman Trend. I am interested to know whether the number of sample is sufficient from the power for the sample size of this parameter. However I am a bit confused as after doing some study I found that I need two group where the difference in mean of the two groups are subtracted and divided by the standard deviation to find the effect size. But in this case there is only one group i.e. the data which is used for the computation of trend. Therefore my question is given the sample size and the data from which I can compute the mean and standard deviation, how do I find the effect size and compute the power to identify whether the number of sample is sufficient for the trend analysis. Can anyone help?

  2. #2
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,185 Times in 1,146 Posts

    Re: Identifying Effect size in Power Analysis

    To have a trend you need 2 variables, so how don't you have two variables?


    Water variable against time?
    Stop cowardice, ban guns!

  3. #3
    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: Identifying Effect size in Power Analysis

    Hi,
    for a Spearman correlation the effect size is the smallest value of the correlation coefficient that you want to detect with a given probability, so you just need to decide on that, you do not need two groups etc.

    regards

  4. #4
    Omega Contributor
    Points: 38,334, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,185 Times in 1,146 Posts

    Re: Identifying Effect size in Power Analysis

    So you don't need two groups, but you need two variables, this is where I was getting confused by the OP?
    Stop cowardice, ban guns!

  5. #5
    Points: 3,628, Level: 37
    Level completed: 86%, Points required for next Level: 22

    Posts
    11
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Identifying Effect size in Power Analysis

    Quote Originally Posted by hlsmith View Post
    To have a trend you need 2 variables, so how don't you have two variables?


    Water variable against time?
    Yes in that sense there are two groups. However comparing the average of time vs the average of water quality variable does not make much sense...

  6. #6
    Points: 3,628, Level: 37
    Level completed: 86%, Points required for next Level: 22

    Posts
    11
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Identifying Effect size in Power Analysis


    Quote Originally Posted by rogojel View Post
    Hi,
    for a Spearman correlation the effect size is the smallest value of the correlation coefficient that you want to detect with a given probability, so you just need to decide on that, you do not need two groups etc.

    regards
    Re: Identifying Effect size in Power Analysis



    Can you please elaborate a little further with an example. Lets say you have five set of data for Aluminum collected from 5 water samples:
    Date Aluminum (mg/L)
    Jan 1, 2016 0.0013
    Feb 1, 2016 0.0015
    Mar 1, 2016 0.0012
    April 1, 2016 0.05
    May 1, 2016 0.003

    So is this about the smallest detectable size which I am deciding on aluminum perhaps based on historic maximum and minimum?

+ 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