+ Reply to Thread
Results 1 to 3 of 3

Thread: Violating the assumption of normality and homogeneity of variances. How to report?

  1. #1
    Points: 52, Level: 1
    Level completed: 4%, Points required for next Level: 48

    Posts
    4
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Violating the assumption of normality and homogeneity of variances. How to report?




    I am trying to run One way ANOVA but after trying to test these two assumptions, I found that the data violated the assumptions.

    I am now trying to use the non-parametric Kruskal wallis H test instead. But I have problems in reporting the violation of the assumptions in APA style.

    Below are the results of the tests
    Tests of Normality:
    Shapiro-Wilk:
    Group 1: df = 140 p = .096
    Group 2: df = 182 p = .001
    Group 3: df = 59 p = .124

    Test of homogeneity of variances
    Levene statistic = 3.02 df1 = 2 df2 = 378 p = .021


    It will be nice if you guys can suggest other methods to deal with these violations of assumptions too!
    Thanks a bunch!

  2. #2
    TS Contributor
    Points: 17,779, Level: 84
    Level completed: 86%, Points required for next Level: 71
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,542
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Violating the assumption of normality and homogeneity of variances. How to report

    With n=384, violation of the normality assumption does not affect the
    ANOVA results. Regarding the inhomogenous variances, there are
    corrections for this (Brown-Forsythe, Welch).

    With kind regards

    K.

  3. The Following 2 Users Say Thank You to Karabiner For This Useful Post:

    Nivv (12-06-2015), rogojel (12-06-2015)

  4. #3
    Points: 52, Level: 1
    Level completed: 4%, Points required for next Level: 48

    Posts
    4
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Re: Violating the assumption of normality and homogeneity of variances. How to report


    I see.
    So my Welch test gives significant results.
    I have found that there are post hoc tests for unequal variances such as the Games-Howell post hoc test, so proceeding with that will be appropriate?
    Last edited by Nivv; 12-06-2015 at 09:04 AM.

+ 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