1

Which is the appropriate test to test normality if there are only 5 samples? I know that Shapiro-Wilk test doesn't work for small size samples. Is Kolmogorov-Smirnov test appropriate?

Thank you

Xing
  • 11
  • 1
  • 1
    Look at this - http://stats.stackexchange.com/questions/13983/is-it-meaningful-to-test-for-normality-with-a-very-small-sample-size-e-g-n – Dark_Knight Nov 14 '16 at 15:45
  • 1
    A Shapiro-Wilk can be done with $n=3$. In what sense do you mean "doesn't work" and how do you know it? – Glen_b Nov 14 '16 at 16:21
  • 1
    You might also like to expand on why you wish to test for normality. – mdewey Nov 14 '16 at 16:29

0 Answers0