[R] histogram not equal to Kolmogorov-Smirnov test; 3-way-nested Anova (2)

[David Winsemius]

On Aug 8, 2011, at 10:00 AM, Jörg Stephan wrote:

> Hello R-Help-Team,
>
> I am doing a 3-way-nestedAnova and a very strange thing occurred. My  
> "Two-sample Kolmogorov-Smirnov test" (p=0.957) meet the requirement  
> of normal distribution, BUT if I have a look at the histogram it is  
> definitely not normally distributed. I never had something like that  
> and I am very confused. I attached the data and my used code; in  
> case you are not able to open it or need further information?s  
> please tell me. As it is my first nested Anova, it would be very  
> nice if you (hopefully) confirm the used code. Hoping you can help  
> me. Thank you very much in advance.

Is this homework? Homework submissions are discouraged on this list.  
Looking at marginal distributions and then making inferences about  
residual seems like a student error.

-- 

David Winsemius, MD
West Hartford, CT