Normality test

#1
I will use Normal probability plot or Shapiro-Wilk test for normality check. But is it that straightforward or there is some subtlety in this?

Many Thanks
 

Karabiner

TS Contributor
#2
The question was whether the given data are normally distributed. Therefore, a test of significance doesn't apply, because it tests whether the assumption can be rejected that these data were sampled from a normally distributed population.

Moreover, even if ohne wanted a test of significance here, a non-significant result would not mean much, because power to reject the null hypothesis (= normality assumption for the population) is very low, due to small sample size(s).

With Kind regards

K.