+ Reply to Thread
Results 1 to 3 of 3

Thread: Help choosing right statistics

  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

    Help choosing right statistics




    The experiment consisted of testing the effect of several diets on the weight of fish. I want to check what groups differ from each other. There is no homogeneity of variance between groups. Is it correct if i use the Kruskall Wallis test (I have read that the kruskall-wallis should not be used when the variances are different)? I know this test only answer the question " is there a difference" and not " where is the difference". What test can i do after to check where are those difference?
    I'm confused with the amount of information i have found.. It sometimes seem it is more a matter of preference than of being right or wrong..
    If someone could help me i would be greatly appreciated. I can give more details of the experiment if necessary.
    Thanks in advance,
    André Barreto

  2. #2
    TS Contributor
    Points: 17,949, Level: 85
    Level completed: 20%, Points required for next Level: 401
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,561
    Thanks
    56
    Thanked 644 Times in 606 Posts

    Re: Help choosing right statistics

    How many groups are there, how large is your total sample size, are sample sizes different between groups? How large is the smallest standard deviation and how large is the largest standard deviation within groups?

    With kind regards

    Karabiner
    »Jetzt kann mich der Führer mal am Arsch lecken.« (Ernst Kuzorra, 1941)

  3. #3
    Omega Contributor
    Points: 39,072, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,075
    Thanks
    402
    Thanked 1,192 Times in 1,153 Posts

    Re: Help choosing right statistics


    Did you randomize the diet or can you rule out needing to control for confounders?


    Why not use linear regression analysis?
    Stop cowardice, ban guns!

+ 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