Re: Basic rookie question about standard error
- From: "Kevin E. Thorpe" <kevin.thorpe@xxxxxxxxxxx>
- Date: Fri, 19 Aug 2005 14:16:06 GMT
BillCo wrote:
> Thanks for all the help guys, that really clears thing up for me at
> last!!
>
> The zeros do not represent n/a and hence they should be used for
> calcualtion of the mean and standard error :)
One question. Why are you calculating SE? If the purpose is to
describe the variability of the observed data, you should
calculate the SD. The SE is telling you about the variability
of the sampling distribution of the mean.
--
Kevin E. Thorpe
Biostatistician/Trialist, Knowledge Translation Program
Assistant Professor, Department of Public Health Sciences
Faculty of Medicine, University of Toronto
.
- Follow-Ups:
- Re: Basic rookie question about standard error
- From: BillCo
- Re: Basic rookie question about standard error
- References:
- Basic rookie question about standard error
- From: coleman . bill
- Re: Basic rookie question about standard error
- From: Paige Miller
- Re: Basic rookie question about standard error
- From: Richard Ulrich
- Re: Basic rookie question about standard error
- From: Data Matter
- Re: Basic rookie question about standard error
- From: BillCo
- Basic rookie question about standard error
- Prev by Date: Re: Basic rookie question about standard error
- Next by Date: Detecting overdispersion in Poisson regression
- Previous by thread: Re: Basic rookie question about standard error
- Next by thread: Re: Basic rookie question about standard error
- Index(es):