+ Reply to Thread
Results 1 to 8 of 8

Thread: Parametric vs nonparametric tests

  1. #1
    Points: 29, Level: 1
    Level completed: 58%, Points required for next Level: 21

    Posts
    3
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Parametric vs nonparametric tests




    I'm pretty confused about the conditions in which I should use a parametric or a nonparametric test to compare the means of 2 (or more) populations and would be very grateful if someone would explain it to me.

    I know that if the distribution of the sample is not normal, non-parametric (Wilcoxon, Kruskal-Wallis) should be used. It is assumed that the central limit theorem allows to use parametric if n> 30, generalized error because in any case they are n> 30 for each level of the factor. On the other hand, now it seems that CLT is questioned and that 30 cases does not guarantee normality.

    When do we have to use one or the other?

    If ANOVA requires homoscedasticity, why are post-hoc tests for unequal variances?

    Many thanks

  2. #2
    TS Contributor
    Points: 18,889, Level: 87
    Level completed: 8%, Points required for next Level: 461
    CowboyBear's Avatar
    Location
    New Zealand
    Posts
    2,062
    Thanks
    121
    Thanked 427 Times in 328 Posts

    Re: Parametric vs nonparametric tests

    Quote Originally Posted by david030177 View Post
    I know that if the distribution of the sample is not normal, non-parametric (Wilcoxon, Kruskal-Wallis) should be used.
    A few quick thoughts on this:

    1) It is the distributions of the errors (i.e., the distributions within each groups) that are assumed normal, not the marginal distribution across all groups

    2) Unless you have a tiny sample size and the departure from normality is very extreme, a breach of normality is unlikely to cause meaningful problems. I blogged about this recently. Normality is an assumption of these tests, but it is by far the least important of several assumptions.

    3) If you are worried about non-normality, bootstrapping or permutation tests allow you to test for mean differences without changing your whole framework of inference (Note: A Wilcoxon or Kruskal-Wallis test is not a test for differences in means, or at least not unless we assume that the distributions have the same shape and spread within each sample; without that assumption these tests have quite different and counterintuitive null hypotheses).

    PS. I see this is your first post.
    Matt aka CB | twitter.com/matthewmatix

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

    david030177 (11-18-2016)

  4. #3
    Points: 29, Level: 1
    Level completed: 58%, Points required for next Level: 21

    Posts
    3
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Re: Parametric vs nonparametric tests

    MANY THANKS CowboyBear for your quick reply !!
    Only another question. How could I know if the distributions of both samples are similar?
    Thank you again
    Best wishes
    David

  5. #4
    Points: 358, Level: 7
    Level completed: 16%, Points required for next Level: 42

    Posts
    15
    Thanks
    0
    Thanked 3 Times in 3 Posts

    Re: Parametric vs nonparametric tests

    To know if your distributions are normal you can conduct tests of normality such as the Kolmogorov-Smirnov test (K-S test).

  6. The Following User Says Thank You to gdaem For This Useful Post:

    david030177 (11-18-2016)

  7. #5
    Points: 29, Level: 1
    Level completed: 58%, Points required for next Level: 21

    Posts
    3
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Re: Parametric vs nonparametric tests

    Thank you gdaem, but I asked for the shape of two samples to compare the means with a nonparametric test, if I understood correctly CowboyBear's reply. Not necessarily to check if they are normal distributions.
    Best wishes
    David

  8. #6
    Super Moderator
    Points: 13,151, Level: 74
    Level completed: 76%, Points required for next Level: 99
    Dragan's Avatar
    Location
    Illinois, US
    Posts
    2,014
    Thanks
    0
    Thanked 223 Times in 192 Posts

    Re: Parametric vs nonparametric tests

    Quote Originally Posted by david030177 View Post
    I'm pretty confused about the conditions in which I should use a parametric or a nonparametric test to compare the means of 2 (or more) populations and would be very grateful if someone would explain it to me.

    I know that if the distribution of the sample is not normal, non-parametric (Wilcoxon, Kruskal-Wallis) should be used. It is assumed that the central limit theorem allows to use parametric if n> 30, generalized error because in any case they are n> 30 for each level of the factor. On the other hand, now it seems that CLT is questioned and that 30 cases does not guarantee normality.

    When do we have to use one or the other?

    If ANOVA requires homoscedasticity, why are post-hoc tests for unequal variances?

    Many thanks

    What you are forgetting is the issue of Power versus Type I error. That is, nonparametric rank-based tests can demonstrate to be much more powerful than the usual parametric (t or F) tests when the underlying distributions are skewed and/or heavy-tailed....Studies abound.

  9. The Following User Says Thank You to Dragan For This Useful Post:

    david030177 (11-19-2016)

  10. #7
    TS Contributor
    Points: 40,089, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Downloads
    gianmarco's Avatar
    Location
    Italy
    Posts
    1,367
    Thanks
    232
    Thanked 301 Times in 225 Posts

    Re: Parametric vs nonparametric tests

    You could perform a permutation t-test. Besides obtaining a p value, you may want to inspect the distribution of the permuted mean differences. If it heavily non-normal, it indicates that regular t-test would have not worked fine.
    I have put togheter a R function to do that. It is available from:
    http://cainarchaeology.weebly.com/r-...on-t-test.html

    Hope this helps
    Best
    http://cainarchaeology.weebly.com/

  11. The Following User Says Thank You to gianmarco For This Useful Post:

    david030177 (11-19-2016)

  12. #8
    TS Contributor
    Points: 18,889, Level: 87
    Level completed: 8%, Points required for next Level: 461
    CowboyBear's Avatar
    Location
    New Zealand
    Posts
    2,062
    Thanks
    121
    Thanked 427 Times in 328 Posts

    Re: Parametric vs nonparametric tests


    Quote Originally Posted by david030177 View Post
    MANY THANKS CowboyBear for your quick reply !!
    Only another question. How could I know if the distributions of both samples are similar?
    Thank you again
    Best wishes
    David
    Shape: You could use histograms or a qqplot (qqplot of the one data distribution against the other, instead of one data distribution against a normal distribution, as in a qqnorm plot).

    Spread: Levene's test or just look at the sample variances.

    But making analytic decisions contingent on data can be problematic, so if you do decide to do a different analysis than you had initially planned, do report both.
    Matt aka CB | twitter.com/matthewmatix

  13. The Following User Says Thank You to CowboyBear For This Useful Post:

    david030177 (11-19-2016)

+ 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