+ Reply to Thread
Results 1 to 3 of 3

Thread: p value for pooled data

  1. #1
    Points: 15, Level: 1
    Level completed: 29%, Points required for next Level: 35

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    p value for pooled data




    Hi all,

    In my research analysis, I've used the multiple imputation technique in SPSS to handle missing data. So far, so good.

    Now I'm running the analyses, that are all non-parametric (Mann-Whitney U and Wilcoxon signed ranks test). I've noticed that I get test statistics for my original data and for each of my imputation step (10), but not for my pooled data, although I get ranks for the pooled data.

    I need a p value for my pooled data, since that is the test statistic for my dataset after multiple imputation.

    Is there any way to solve my problem?

  2. #2
    Omega Contributor
    Points: 38,289, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    6,992
    Thanks
    397
    Thanked 1,185 Times in 1,146 Posts

    Re: p value for pooled data

    Not that familiar with SPSS, can you just create a new dataset consisting of your full sample (with the MI data), then run the tests?
    Stop cowardice, ban guns!

  3. #3
    Points: 15, Level: 1
    Level completed: 29%, Points required for next Level: 35

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: p value for pooled data


    I think that's not exactly what pooled data consist of. When I run an analysis leaving the original data out and not handling the dataset as containing multiple imputation data (i.e. not split by imputation number), than you'll have a (very) large n, so the statistics are troubled.

+ 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