[R] Question

csrabak crabak at acm.org
Thu Dec 16 16:42:38 CET 2010


Em 14/12/2010 12:46, Matthew Rosett escreveu:
> How do I determine if my data deviate from the normal distribution?
> The sample size is 1000 (weights of people).
>
As others already posted about the Statistical theoretic aspects of it, 
I want to add that weights of people are not normal distributed due 
biological/physical reasons:

There is a minimum weight an individual below no one can surpass; same 
for maximum.

Also the probabilities for symmetrical z scores are not the same, etc.

This lead researchers in the area of anthropometrics to create 
transformations on these distributions in order to be able to use test 
we're are used to with normal distributions.

See for an example: http://www.jstor.org/pss/2982992.

--
Cesar Rabak



More information about the R-help mailing list