+ Reply to Thread
Results 1 to 2 of 2

Thread: Doing Parametric Tests without normal distribution

  1. #1
    Points: 271, Level: 5
    Level completed: 43%, Points required for next Level: 29

    Posts
    10
    Thanks
    9
    Thanked 0 Times in 0 Posts

    Doing Parametric Tests without normal distribution




    Hello everyone!
    Is it true that when your sample is more than 30 people, you can do parametric tests, even though your sample may not follow the normal distribution?

    In my research I have 400 people as a sample, not following normal distribution and someone told me that I can do parametric tests anyway... But I wanted to ask the experts (you) to confirm that in case this is not true.

    If someone knows something about that, please(!!!!!!) let me know! Thank you in advance for your time!!

  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: Doing Parametric Tests without normal distribution


    You need not care for the distribution of your dependent variable anyway, regardless of n > 30 or n <30. Instead, the residuals of your model (e.g. t-test, ANOVA, linear regression) should preferably come from a normal distribution. But it is true that with n > 30 one can assume that the tests are valid, even if the resiudals are not from a normal distribution. You can look for the central limit theorem.

    With kind regards

    Karabiner

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

    Curtains White (02-12-2017)

+ Reply to Thread

           




Tags for this 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