+ Reply to Thread
Results 1 to 3 of 3

Thread: What to do with unequal sample sizes?

  1. #1
    Points: 58, Level: 1
    Level completed: 16%, Points required for next Level: 42

    Posts
    6
    Thanks
    2
    Thanked 0 Times in 0 Posts

    What to do with unequal sample sizes?




    I am working on this time-effect project in which we pulled data at different time points from a big old data set. The issue is that all subjects were sacrificed at different times after treatment. So I have 16 samples at time 1, 10 samples at time 2, and so on. I am going to do Friedman test and one-way repeated measure ANOVA on those data (one for parametric and one for nonparametric). But I don't know what to do with the unequal sample size. Since my smallest group only has 4 values, SPSS only count 4 from each group when doing the analysis.

    I've read about a test called Skillings-Mack Test that can be an alternative to Friedman test in this situation. But it's not available in SPSS. I've also seen something about using linear mixed model to do repeated measure ANOVA. But it seems to be a little complicated for me.

    So, any advice?

  2. #2
    Omega Contributor
    Points: 38,374, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: What to do with unequal sample sizes?

    Haven't heard of S-M, but know that multilevel models (mixed) can handle these types of data. And yes, they are very difficult.
    Stop cowardice, ban guns!

  3. The Following User Says Thank You to hlsmith For This Useful Post:

    skylandyu (07-21-2015)

  4. #3
    Omega Contributor
    Points: 38,374, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,998
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: What to do with unequal sample sizes?


    Just skimmed the M-S test. The first description I saw said it can handle arbitrary missing data, that means missing at any time point. You appear to have monotonic missing data (kind of like attrition). Check this out before proceeding with M-S.
    Stop cowardice, ban guns!

  5. The Following User Says Thank You to hlsmith For This Useful Post:

    skylandyu (07-21-2015)

+ Reply to Thread

           




Tags for this 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