+ Reply to Thread
Results 1 to 2 of 2

Thread: Is this correct?- Kriskal Wallis Test

  1. #1
    Points: 32, Level: 1
    Level completed: 64%, Points required for next Level: 18

    Thanked 0 Times in 0 Posts

    Is this correct?- Kriskal Wallis Test

    I want to compare multiple groups pre-intervention and post-intervention, to see which intervention was the most effective.
    So I calculated the Change (postintervention minus preintervention) and used this as the testing variable. The grouping variable I used as the actual intervention itself.
    It gave me each intervention with mean ranks.

    Does this sound okay? Also, does the mean ranks just show that there is a difference- or does it also show the ranked difference (i.e. higher mean-ranked performed better than lower mean-ranked)?


  2. #2
    TS Contributor
    Points: 14,811, Level: 78
    Level completed: 91%, Points required for next Level: 39
    Miner's Avatar
    Greater Milwaukee area
    Thanked 405 Times in 363 Posts

    Re: Is this correct?- Kriskal Wallis Test

    Is there a reason that you do not use a 1-way ANOVA on the Change variable using intervention as the treatment levels? ANOVA is robust against considerable departure from normality. You can evaluate the residuals to verify that the assumptions have been met. You can then perform a Post-Hoc multiple comparison of means to determine the most effective. See Hsu's Multiple Comparison's with the Best (MCB) test.

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

    03turnerS (04-11-2013)

+ 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