[R] normality test

[Frank E Harrell Jr]

Romain Francois wrote:
> Le 28.04.2005 13:16, Pieter Provoost a écrit :
> 
>> Hi,
>>
>> I have a small set of data on which I have tried some normality tests. 
>> When I make a histogram of the data the distribution doesn't seem to 
>> be normal at all (rather lognormal), but still no matter what test I 
>> use (Shapiro, Anderson-Darling,...) it returns a very small p value 
>> (which as far as I know means that the distribution is normal). 
>> Am I doing something wrong here?
>> Thanks
>> Pieter
>>  
>>
> Hello,
> 
> You seem to know not far enougth.
> Null hypothesis in shapiro.test is **normality**, if your p-value is 
> very small, then the data is **not** normal.
> 
> Look carefully at ?shapiro.test and try again. Furthermore, normality 
> tests are not very powerful. Consider using a ?qqnorm and ?qqline
> 
> Romain
> 

Usually (but not always) doing tests of normality reflect a lack of 
understanding of the power of rank tests, and an assumption of high 
power for the tests (qq plots don't always help with that because of 
their subjectivity).  When possible it's good to choose a robust method. 
  Also, doing pre-testing for normality can affect the type I error of 
the overall analysis.

-- 
Frank E Harrell Jr   Professor and Chair           School of Medicine
                      Department of Biostatistics   Vanderbilt University