+ Reply to Thread
Results 1 to 3 of 3

Thread: Pre and post comparison for nonparametric unpaired sample

  1. #1
    Points: 4, Level: 1
    Level completed: 7%, Points required for next Level: 46

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Pre and post comparison for nonparametric unpaired sample




    Hi. I am not sure how to compare the results from the following study.
    Number of wrong orders out of total orders placed every day for 60 days as the pre sample (significant variation eg 2 wrong out of 6 on day 1, 5 wrong out of 15 on day 2, and so on till day 60)
    Intervention to decrease the number of wrong orders placed
    Similar data collected for next 60 days

    What test do I use? Mann Whitney? Unpaired t test? Some other test? How do I analyze this data?
    Thanks for your help

  2. #2
    Points: 1,974, Level: 26
    Level completed: 74%, Points required for next Level: 26

    Location
    New Zealand
    Posts
    227
    Thanks
    3
    Thanked 48 Times in 47 Posts

    Re: Pre and post comparison for nonparametric unpaired sample

    One suggestion is to accumulate the data before and after. e.g. Before 125 right out of 180, After 155 right out of 175. Then use the Chi Square test, or test for the difference between two proportions.

  3. #3
    TS Contributor
    Points: 14,811, Level: 78
    Level completed: 91%, Points required for next Level: 39
    Miner's Avatar
    Location
    Greater Milwaukee area
    Posts
    1,171
    Thanks
    34
    Thanked 405 Times in 363 Posts

    Re: Pre and post comparison for nonparametric unpaired sample


    The following suggestion may not be accepted in the biostatistics field, but is common in my field of industrial statistics.

    Since your samples are taken in time series order, a common tool in my field would be a p-chart. See example below. This is a general explanation of what a p-chart is and what it is used to monitor. The second image are tests to detect different types of instability. The one of most interest to you would be the 9 points in a row on the same side of the center line. The center or mean line is calculated from the first 60 days data.

    Another approach would be to use ANOM. While ANOM and the suggested chi-square are legitimate approaches, they both ignore the information inherent in the time series structure.
    Attached Thumbnails Attached Thumbnails Click image for larger version

Name:	p chart rules.jpg‎
Views:	15
Size:	49.2 KB
ID:	6542   Click image for larger version

Name:	p chart.jpg‎
Views:	13
Size:	31.2 KB
ID:	6541  

+ 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