+ Reply to Thread
Results 1 to 2 of 2

Thread: Mann-Whitney Non-parametric tests

  1. 07-22-2008 06:15 PM #1
    urtiga
    urtiga is offline
    Points: 2,952, Level: 33
    Level completed: 35%, Points required for next Level: 98
    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Smile Mann-Whitney Non-parametric tests



    Hi,

    I need to perform non-parametric tests:Mann-Whitney. But, I've read that to perform this test, I need to assure the equality of variances (levene test) and the equality of the distribution (boxplots, etc)...but, I couldn't assure this assumptions...this mean I cant use this non-parametric test? In such case, what can I do?
    Thanks
    Reply With Quote Reply With Quote
  2. 07-22-2008 06:21 PM #2
    jamesmartinn
    jamesmartinn is offline
    Points: 9,065, Level: 64
    Level completed: 5%, Points required for next Level: 285
    Posts
    482
    Thanks
    22
    Thanked 3 Times in 3 Posts

    You do not need equality of variances for the Mann Whitney
    Reply With Quote Reply With Quote
+ Reply to Thread
« Confused about p-values with regression models | Effect size in meta-analysis »

Similar Threads

  1. Mann-whitney
    By dkhamoo in forum Statistics
    Replies: 0
    Last Post: 04-16-2011, 05:31 AM
  2. Should I use Mann-Whitney?
    By GrungyGoose in forum Statistics
    Replies: 7
    Last Post: 02-22-2011, 06:02 AM
  3. t-tests or mann-whitney??
    By zxcvbnm in forum Statistical Research
    Replies: 2
    Last Post: 05-03-2009, 06:41 PM
  4. Mann Whitney
    By jamesmartinn in forum Statistics
    Replies: 0
    Last Post: 03-07-2009, 03:49 PM
  5. Power Analysis for Mann Whitney U and/or variance ratio F tests
    By boomslang in forum Statistics
    Replies: 1
    Last Post: 10-14-2007, 01:57 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts

Forum Rules








Advertise on Talk Stats