bstrap - Obtain and plot bootstrap sampling distributions of the sample
mean
Many common tests assume sufficient normality of the data
distribution, with the received wisdom that if the sample size
is "large", the normality doesn't matter so much (because of
the Central Limit Theorem). It is difficult to judge the
normality is good enough, or whether the sample size is big
enough. A better way to investigate is to obtain a bootstrap
sampling distribution of the sample mean (by taking repeated
bootstrap samples), and to assess that distribution for
normality. If it is, the normal-theory test will work; if not,
not.
