+ Reply to Thread
Results 1 to 3 of 3

Thread: Advice on appropriate statistical test

  1. #1

    Advice on appropriate statistical test




    I have 4 groups who provided scores and am interested in comparing the scores between groups to determine if they are significantly different. The data is not normally distributed so I assume I cannot use ANOVA.

    The 4 groups include 5 individuals per group (20 individuals total). Each individual provided answers to 9 similarly scored questions. In total, I have 45 numbers per group or 180 answers across all 4 groups.

    What statistical test should I use to compare the means between 4 groups? Would my n per group be 45 as I have that many unique answers or would it be 5 as each of the 5 individuals answered 9 questions?

    Thank you for your advice!!

  2. #2
    Points: 3,631, Level: 37
    Level completed: 88%, Points required for next Level: 19
    staassis's Avatar
    Location
    New York
    Posts
    226
    Thanks
    2
    Thanked 41 Times in 39 Posts

    Re: Advice on appropriate statistical test

    You should aggregate 9 questions into just a few subscales. If all 9 questions are very different, ok, you will have to analyze them separately.

    In each group you have only 5 independent observations - 1 for each person. Since the data are non-normal and scarce, you should compare the median of each characteristic (a subscale or an original question) over the 4 groups using a Kruskal-Wallis test.

  3. #3
    Points: 2, Level: 1
    Level completed: 3%, Points required for next Level: 48

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Advice on appropriate statistical test


    Have a look at: Mood’s median test; Kruskal-Wallis test

    they aren't sensitive to the distribution.

+ 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